{"source":"AI Sandbox Hunt","url":"https://aisandboxhunt.com","license":"Free to use with a link to the source page.","updated":"2026-11-15","method":"https://aisandboxhunt.com/methodology","methodVersion":"1.3 (2026-11-14)","categories":[{"id":"code-sandbox","url":"https://aisandboxhunt.com/code-sandboxes","name":"AI agent sandboxes","priceUnit":"workload","workload":"2 vCPU, 4 GiB, 176 hours a month (8 h × 22 days), CPU busy 30%"},{"id":"dev-box","url":"https://aisandboxhunt.com/dev-boxes","name":"Dev boxes for AI agents","priceUnit":"workload","workload":"2 vCPU, 4 GiB, 220 hours a month (10 h × 22 days), CPU busy 20%, paused the rest of the time"},{"id":"browser","url":"https://aisandboxhunt.com/browser-sandboxes","name":"Browser sandboxes for AI agents","priceUnit":"browserHour","workload":null},{"id":"gpu","url":"https://aisandboxhunt.com/gpu-sandboxes","name":"GPU sandboxes","priceUnit":"h100","workload":null},{"id":"ci-runner","url":"https://aisandboxhunt.com/ci-runners","name":"CI runners","priceUnit":"workload","workload":"2 vCPU Linux runner, 100 hours a month"},{"id":"coding-agent","url":"https://aisandboxhunt.com/coding-agents","name":"Coding agents with cloud machines","priceUnit":"none","workload":null}],"providers":[{"slug":"aether","name":"Aether","company":"Aether","website":"https://www.runaether.dev","docsUrl":"https://docs.runaether.dev/introduction","pricingUrl":"https://docs.runaether.dev/concepts/pricing-and-tiers","categories":["dev-box","code-sandbox"],"tagline":"A Firecracker VM per coding-agent task, running your own Claude Code or Codex, billed in credits for active VM time.","summary":"Aether runs each coding-agent task in its own Firecracker microVM, an Ubuntu workspace with persistent storage, a terminal, a browser and SSH access. You bring Claude Code, Codex or OpenCode on your own subscription or API key (or buy model calls separately), and Aether bills the VM in credits for active runtime only, per second. Workspaces suspend when idle and resume where they left off. Plans grant monthly credits, and credit packs cost $0.10 per credit.","isolation":"Firecracker microVM per task, not reused across tasks or users","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["CLI","REST API","WebSocket API","MCP"],"regions":"Not stated on the pages checked.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":1,"instance":"Medium: 2 vCPU / 8 GB / 20 GB volume (10 credits per hour at $0.10 per credit)","storageGibMonth":null,"billingIncrement":"per second (credits prorated by the second)","billedWhile":"Only while the workspace VM is active. Idle and suspended workspaces are not billed; a workspace suspends once no task, background activity or interactive connection holds it.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":2,"limits":"20 credits a month (about 2 hours of Medium, or about 4 hours of the Small size listed on the homepage), 1 warm slot, 3 tasks at once.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Starter","monthlyFee":30,"includedUsageUsd":40,"limits":"400 credits a month (about 40 hours of Medium), 2 warm slots, 10 tasks at once.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":75,"includedUsageUsd":200,"limits":"2,000 credits a month (about 200 hours of Medium), 3 warm slots, 25 tasks at once.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":120,"includedUsageUsd":400,"limits":"4,000 credits a month (about 400 hours of Medium), 5 warm slots, 50 tasks at once.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: 20 credits a month at no cost, permanently (about 2 hours of the 2 vCPU / 8 GB Medium size).","notes":["Pricing is in credits. We convert at the credit-pack price of $0.10 per credit (packs of 100 for $10 up to 1,000 for $100, never expire). Medium uses 10 credits per hour = $1.00/hour, Large (4 vCPU / 16 GB) 20 = $2.00/hour, XL (8 vCPU / 32 GB) 40 = $4.00/hour. Each size doubles the previous one.","Included plan credits are converted at the same $0.10 per credit: Free 20 = $2, Starter 400 = $40, Pro 2,000 = $200, Max 4,000 = $400. Bought through a plan, credits cost less: about $0.075 (Starter), $0.0375 (Pro) and $0.03 (Max) each.","There is no billing after the credits run out: task starts are refused until you buy a credit pack or the plan renews. Packs are prepaid top-ups available on every plan.","Conflict between vendor pages: the homepage says 1 credit = 12 minutes on a default Small devbox (1 vCPU, 4 GB), with Medium (2 vCPU, 8 GB) at 2x and Large (4 vCPU, 16 GB) at 4x. The docs list only Medium, Large and XL, say Medium is the default size, and price Medium at 10 credits per hour. The rates agree (Small would be 5 credits = $0.50 per hour); we show Medium because both pages list it.","On 'boat placements' a workspace is metered per resource instead: 1 credit per CPU-hour and 1 credit per GB-hour of memory while it works, nothing while idle (12 credits per hour for Medium at full use). The pages checked do not say when a workspace runs on a boat placement.","Memory is stated in GB; we treat it as GiB.","Mobile testing devices also spend credits: 1 credit a minute for an Android emulator or Gradle build sandbox, 2 for an iOS simulator or Xcode build sandbox.","Model calls through Aether's own inference are charged at provider list price to a separate prepaid balance, not to credits. Calls on your own subscription or key are not billed by Aether.","Lifetime: the homepage says a devbox and its volume are destroyed when the task's pull request merges or closes, or after 48 hours of inactivity. The docs describe suspend on idle and an option to keep a workspace active."]},"bestFor":["Running Claude Code, Codex or OpenCode on your own subscription in a cloud VM you can watch, steer and SSH into.","Task-per-VM agent work tied to GitHub pull requests, Slack, Linear or schedules.","Bursty agent work, since idle and suspended workspaces cost nothing."],"watchOut":["At the $0.10 credit-pack price a 2 vCPU / 8 GB VM costs $1.00 per hour; plan credits are cheaper but expire monthly.","Workspaces are tied to a task and, per the homepage, are deleted when its pull request merges or closes, or after 48 hours of inactivity.","Only three fixed sizes in the docs, up to 8 vCPU / 32 GB; no GPU.","The homepage and docs disagree on the default size and whether a 1 vCPU Small size exists."],"sources":[{"url":"https://www.runaether.dev","quote":"1 credit = 12 minutes on the default Small devbox (1 vCPU, 4GB).Credits meter active runtime only, per second.","checked":"2026-11-14"},{"url":"https://www.runaether.dev","quote":"Need more? Credit packs on any plan, at $0.10 a credit.","checked":"2026-11-14"},{"url":"https://www.runaether.dev","quote":"Medium (2 vCPU, 8GB) = 2× credits. Large (4 vCPU, 16GB) = 4× credits.","checked":"2026-11-14"},{"url":"https://www.runaether.dev","quote":"the machine and its volume are destroyed when your pull request merges or closes, or after 48 hours of inactivity.","checked":"2026-11-14"},{"url":"https://www.runaether.dev","quote":"Each task gets its own Firecracker microVM that is never reused for another task or another user.","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/pricing-and-tiers","quote":"| Starter | $30 | 400 | 2 |","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/pricing-and-tiers","quote":"| Medium | 2 performance CPUs | 8 GB | 20 GB | 10 |","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/pricing-and-tiers","quote":"| `credits_1000` | 1,000 | $100 |","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/pricing-and-tiers","quote":"GPU runtime classes are not available.","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/billing-and-credits","quote":"Medium is the reference rate: **1 credit = 6 minutes of Medium VM active time** (10 credits/hour).","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/billing-and-credits","quote":"On boat placements a workspace is metered instead: 1 credit per CPU-hour and 1 credit per GB-hour of memory while it works, and nothing while it is idle.","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/workspaces","quote":"| `xl` | 8 | performance | 32 GB | 30 GB | 40 |","checked":"2026-11-14"},{"url":"https://docs.runaether.dev/concepts/workspaces","quote":"| `suspending` | Pausing the VM (snapshotting memory/disk). |","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/aether","score":{"dev-box":{"total":60,"rank":25,"of":38},"code-sandbox":{"total":61,"rank":37,"of":63}},"comparablePrice":{"dev-box":95,"code-sandbox":75},"benchmarks":[]},{"slug":"agent37","name":"Agent37 Cloud","company":"Agent37","website":"https://agent37.com","docsUrl":"https://www.agent37.com/docs","pricingUrl":"https://www.agent37.com/pricing","categories":["code-sandbox","dev-box"],"tagline":"Persistent agent sandboxes billed per minute from a prepaid wallet, with optional full VMs and auto-sleep.","summary":"Agent37 Cloud runs persistent instances for code, services and agent harnesses, managed through a REST API with a Python client. Default instances are gVisor-isolated containers on shared hosts in the United States; VM instances (own kernel, KVM) cost 2.5x the compute rate and also run in France and India. Rates are published per 730-hour month ($0.80 per vCPU, $0.70 per GB of RAM, $0.09 per GB of disk) and metered per minute: compute while running, disk while the instance is kept. There is no subscription; usage draws down a prepaid balance.","isolation":"gVisor per default instance; optional VM instances with their own kernel and KVM hardware virtualization","isolationKind":"gvisor","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["REST API","Python"],"regions":"Default instances run in the United States. VM instances run in us, eu (France) and ap-south (India, Mumbai); outside the US only VM instances are available.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":4,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.0010958904109589042,"gibHour":0.0009589041095890411,"instanceHour":null,"instance":null,"storageGibMonth":0.09,"billingIncrement":"per minute","billedWhile":"Compute (vCPU and RAM) for the whole time an instance is running, including starting and other transitional states. Sleeping or stopped instances bill disk only.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Prepaid wallet","monthlyFee":0,"includedUsageUsd":0,"limits":"No subscription or minimum; top-ups of $5 to $10,000. The limit is the memory running at once: 32 GB before the first top-up, 40 GB after it, 200 GB once top-ups total $50, 800 GB at $250 and 3,200 GB at $1,000. Sleeping and stopped instances do not count. Max 4 vCPU and 32 GB RAM per instance.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"One-time $5 starter credit after adding a card (the card is not charged), for new accounts; up to 3 instances running at once until the first top-up.","notes":["Rates are published per 730-hour month: $0.80 per vCPU, $0.70 per GB of RAM and $0.09 per GB of disk. We divide the vCPU and RAM rates by 730 to get hourly figures; the disk figure is shown per month as published.","These are default-instance rates. VM instances bill compute at 2.5x and Performance instances (dedicated cores) at 4x; disk is unchanged. Default instances share host hardware, so sustained heavy CPU use may be limited when the host is busy.","RAM is stated in GB without saying GB or GiB; we treat GB as GiB.","Shapes are fixed: 2 vCPU / 4 GB, 4 / 8, 4 / 16 and 4 / 32. 4 vCPU is the ceiling on every type; a request for more is provisioned at 4 vCPU.","Auto-sleep is off by default; when on, an instance sleeps after an idle timeout (default 900 seconds, from 5 minutes to a day) and wakes on the next request. In-memory state usually survives a wake but is not guaranteed. Instances parked in cold storage bill $0.03 per GB-month on compressed bytes stored.","Creating an instance requires one day of its running rate in the wallet. Managed models, search and integrations are billed separately from the same wallet.","Instances persist until deleted and nothing times out mid-task. If the wallet goes negative and automatic top-up cannot charge, instances are suspended; an instance left suspended for 30 days is permanently deleted.","Backups, restore and fork are offered; we count these as snapshots. No GPU option is described."]},"bestFor":["One persistent instance per end user or agent that sits idle most of the day, using auto-sleep to pay mostly for disk.","Hosting agent harnesses such as Claude Code, Codex or OpenCode as always-on instances.","Small, steady workloads on a prepaid balance with no plan fee."],"watchOut":["Default instances are gVisor containers on shared hosts; full VMs cost 2.5x compute.","Maximum 4 vCPU per instance on every type.","Default instances run only in the United States; other regions require VM instances.","Running instances at once are capped by total top-ups, starting at 3."],"sources":[{"url":"https://www.agent37.com/pricing","quote":"Agent37 charges $0.80 per vCPU, $0.70 per GB of RAM and $0.09 per GB of disk, per730-hour month, metered by the minute.","checked":"2026-11-15"},{"url":"https://www.agent37.com/pricing","quote":"A standard 2 vCPU, 4 GB RAM sandbox with 4 GB persistent disk costs $4.76 for 730 running hours.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/billing","quote":"Compute is metered per minute: time an instance spends running bills its full rate, time spent stopped or sleeping bills its disk alone.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/billing","quote":"Add a card from the billing page (nothing is charged) and your workspace gets a one-time $5 credit","checked":"2026-11-15"},{"url":"https://www.agent37.com/pricing","quote":"Performance runs on cores held for it and costs 4x compute.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/billing","quote":"Parked in cold storage | $0.03 per GB per month, on the actual compressed bytes stored.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/instance-sizing","quote":"4 vCPU is the ceiling on every instance type.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/vm-instances","quote":"The same four shapes, and compute meters at 2.5x the default rate","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/vm-instances","quote":"VM instances run in every region: `us`, `eu` (France), and `ap-south` (India, Mumbai).","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/concepts","quote":"Runs on dedicated servers in the United States.","checked":"2026-11-15"},{"url":"https://agent37.com","quote":"Every sandbox is enforced by gVisor, with its own filesystem and a hard disk quota.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/agents-api/instance-auto-sleep","quote":"In-memory state usually survives a wake, but it is not guaranteed: treat anything that must outlive a sleep as a file.","checked":"2026-11-15"},{"url":"https://www.agent37.com/docs/llms.txt","quote":"Cut an instance off from the internet, or let it reach only the domains and addresses you list.","checked":"2026-11-15"},{"url":"https://www.agent37.com/pricing","quote":"The limit is the memory running at once: 32 GB before your first top-up, which is 8 sandboxes at 2 vCPU and 4 GB.","checked":"2026-11-15"},{"url":"https://www.agent37.com/pricing","quote":"It rises to 40 GB after your first top-up, 200 GB once your top-ups total $50, 800 GB","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/agent37","score":{"code-sandbox":{"total":70,"rank":16,"of":63},"dev-box":{"total":70,"rank":9,"of":38}},"comparablePrice":{"code-sandbox":1.0608219178082192,"dev-box":1.326027397260274},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":3832,"unit":"ms","better":"lower","date":"2026-11-13","rank":23,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":193271,"unit":"ms","better":"lower","date":"2026-11-13","rank":25,"of":25}]},{"slug":"airtop","name":"Airtop","company":"Airtop","website":"https://www.airtop.ai","docsUrl":"https://docs.airtop.ai","pricingUrl":"https://www.airtop.ai/pricing","categories":["browser"],"tagline":"Cloud browsers with natural-language controls and an agent builder, sold as monthly credit plans.","summary":"Airtop runs cloud browser sessions that agents control through its API with natural-language actions such as page queries and form filling, or directly with Puppeteer, Playwright or Selenium. It also sells Agent Builder, which compiles described workflows into scheduled agents, and Mark, a marketing assistant. All usage, including browser time, proxy bandwidth and AI calls, is paid in credits from a monthly plan, but Airtop does not publish how many credits a browser-hour uses. Plans run from a free tier with 1,000 credits a month to $558 a month.","isolation":"Each session runs in its own encrypted environment; the isolation technology is not stated","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Node.js","Python","REST API","Puppeteer","Playwright","Selenium","n8n","Make","Zapier"],"regions":"Not stated","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not stated","billedWhile":"Credits are charged for the browser's CPU time, proxy bandwidth and AI calls; sessions end after 10 minutes of inactivity by default","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1,000 credits per month; 3 simultaneous sessions; 1 deployed agent. No credit purchases beyond the plan.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Starter","monthlyFee":29,"includedUsageUsd":0,"limits":"30,000 credits per month at the base price (larger allowances up to 150,000 credits available at higher prices); 3 simultaneous sessions; 10 deployed agents; integrated proxy.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Professional","monthlyFee":189,"includedUsageUsd":0,"limits":"225,000 credits per month at the base price (up to 500,000 at higher prices); 30 simultaneous sessions; 30 deployed agents; custom proxy.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Enterprise","monthlyFee":558,"includedUsageUsd":0,"limits":"775,000 credits per month at the base price (up to 1,500,000 at higher prices); 100 simultaneous sessions; unlimited deployed agents; SOC 2 Type 2 report.","includedVcpuHours":0,"includedGibHours":0,"capped":true}],"freeTierKind":"recurring","freeTier":"Free plan with 1,000 credits every month and 3 simultaneous sessions, plus a one-time signup credit bonus in the first month.","notes":["No per-hour browser price is published. Browser CPU time, proxy bandwidth and AI calls all draw from the same credit balance, and the credit cost of a browser-hour is not stated.","Credits reset monthly and do not carry over. One-off credit purchases are not offered; to get more credits you move to a larger plan (upgrades are prorated). Annual billing gives a 10% discount with all credits upfront.","Signup bonus ambiguity: the pricing page lists a 10,000 one-time credit bonus, while the billing help page says new accounts get a one-time grant of 30,000 credits.","An integrated residential proxy is included on paid plans; a custom proxy needs Professional or higher. Captcha solving is offered; its credit cost is not stated.","The site shows a notice that Airtop is working on something new, and its marketing now centres on Agent Builder and Mark; the browser session API remains documented and available."]},"bestFor":["Agents that want natural-language page queries and actions instead of writing selectors","No-code automation builders using n8n, Make or Zapier","Workflows behind logins using saved browser profiles"],"watchOut":["No published price per browser-hour, so costs are hard to compare or predict","No pay-as-you-go overage: when credits run out you must upgrade the plan","Only 3 simultaneous sessions on Free and Starter","The company signals a product change, and its focus has moved toward agent building"],"sources":[{"url":"https://www.airtop.ai/pricing","quote":"10,000 one-time credit bonus","checked":"2026-11-13"},{"url":"https://www.airtop.ai/pricing","quote":"225,000 Credits","checked":"2026-11-13"},{"url":"https://www.airtop.ai/pricing","quote":"100 simultaneous sessions","checked":"2026-11-13"},{"url":"https://www.airtop.ai/pricing","quote":"Airtop is working on something new. Stay tuned!","checked":"2026-11-13"},{"url":"https://www.airtop.ai/help/billing/billing","quote":"The Free plan includes 1,000 credits every month.","checked":"2026-11-13"},{"url":"https://www.airtop.ai/help/billing/billing","quote":"Airtop does not offer one-off credit purchases.","checked":"2026-11-13"},{"url":"https://www.airtop.ai/help/billing/billing","quote":"New accounts also receive a one-time signup grant of 30,000 credits","checked":"2026-11-13"},{"url":"https://docs.airtop.ai/guides/misc/faq","quote":"Airtop also charges credits for CPU time of the browser instance, as well as proxy bandwidth, if you are using a proxy.","checked":"2026-11-13"},{"url":"https://docs.airtop.ai/guides/how-to/creating-a-session","quote":"By default, sessions have an idle timeout of 10 minutes.","checked":"2026-11-13"},{"url":"https://www.airtop.ai/web-automation","quote":"Real browsers that run in the cloud, not on your computer.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/airtop","score":{"browser":{"total":38,"rank":11,"of":12}},"comparablePrice":{"browser":null},"benchmarks":[]},{"slug":"alibaba-agent-sandbox","name":"Alibaba Cloud Agent Sandbox","company":"Alibaba Cloud","website":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/agent-sandbox-overview","docsUrl":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/agent-sandbox-overview","pricingUrl":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","categories":["code-sandbox"],"tagline":"E2B-compatible VM-isolated sandboxes for AI agents on Alibaba Cloud, billed per second with hibernation states.","summary":"Agent Sandbox is Alibaba Cloud's isolated execution environment for AI agents, with code, browser, desktop and mobile sandbox types. The fully managed Flash Edition works with the E2B SDK and CLI, while the Container Edition exposes sandboxes through the Kubernetes API on Container Compute Service. Each sandbox runs in its own VM environment, and a newer MicroVM runtime in beta adds snapshots, forking and Windows. Billing is per second for vCPU, memory and disk above a 15 GiB free allowance, in three compute classes; idle sandboxes can drop into light or deep hibernation, which stops CPU or CPU and memory charges.","isolation":"Each sandbox runs in an independent VM environment; the second-generation runtime (beta) uses lightweight MicroVM isolation","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["E2B SDK (Python, TypeScript)","E2B CLI","Java and Go SDKs maintained by Alibaba Cloud","Kubernetes API (Container Edition)","Alibaba Cloud OpenAPI"],"regions":"Flash Edition: China (Beijing, Ulanqab, Shanghai, Hangzhou, Shenzhen, Hong Kong), Singapore, Malaysia (Johor), US (Virginia) and US (Silicon Valley). Container Edition: over 25 regions across China, Asia-Pacific, Europe, the Americas and the Middle East.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.017856,"gibHour":0.008928,"instanceHour":null,"instance":null,"storageGibMonth":0.1847,"billingIncrement":"per second, 1-second minimum, billed hourly","billedWhile":"While active: vCPU, memory and disk above 15 GiB, on the size set at creation. Light hibernation (Flash Edition only): memory and disk, no vCPU. Deep hibernation and saved snapshots: storage only, for memory size plus disk size. Nothing after the sandbox is destroyed.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"Three compute classes (Economy, Default, Performance) with the same billing rules and different unit prices. Flash Edition default quota of 300 sandboxes per region.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free trial or monthly allowance is stated for Agent Sandbox; only the first 15 GiB of disk on an active sandbox is free.","notes":["Listed rates are the Default (recommended) class for China (Hong Kong) and regions outside China: USD 0.017856 per vCPU-hour and USD 0.008928 per GiB-hour of memory.","Outside mainland China, the Economy class is USD 0.013716 per vCPU-hour and USD 0.006876 per GiB-hour; the Performance class (Flash Edition only, dedicated resource pool) is USD 0.027324 per vCPU-hour and USD 0.013896 per GiB-hour.","Chinese mainland regions are cheaper: the Default class is USD 0.01224 per vCPU-hour and USD 0.006012 per GiB-hour.","Disk is USD 0.00025308 per GiB-hour outside mainland China; per 730-hour month that is about USD 0.1847 per GiB. An active sandbox gets 15 GiB free, but hibernated sandboxes and snapshots pay for all stored data, counted as memory size plus disk size.","Pre-warming custom templates is billed at the disk rate from November 15, 2026; official templates are not charged for pre-warming.","Other Alibaba Cloud services used alongside the sandbox, such as storage or internet egress, are billed by those services.","Pause and resume are available only to allowlisted accounts, and snapshots are an invited test that needs the second-generation runtime; snapshots expire after 7 days by default.","A sandbox can run at most 86,400 seconds (24 hours) at a time in the active state.","Maximum vCPU and memory per sandbox and GPU support for the managed Flash Edition are not stated on the pages checked.","The Flash Edition is closely tied to FC Agent Sandbox in Function Compute, whose documentation describes the same per-second model under Eco, Std and Pro plan names; the Java and Go SDKs are maintained by the Function Compute team."]},"bestFor":["Teams already on Alibaba Cloud, or serving users in China and Asia-Pacific, that need agent sandboxes","Moving E2B SDK code to another provider by changing the endpoint and API key","Long agent sessions that wait on models or people, since hibernation stops CPU and memory charges","Reinforcement learning rollouts and evaluations that clone many sandboxes from one checkpoint"],"watchOut":["Pause and resume need an allowlisted account, and snapshots are an invited test","The MicroVM runtime that adds snapshots and Windows is still beta and only accepts Ubuntu- or Debian-based images","Per-sandbox CPU and memory limits are not published","Prices differ between mainland China and other regions, and the Performance class is only in the Flash Edition"],"sources":[{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/agent-sandbox-overview","quote":"Agent Sandbox is a secure, isolated execution environment where AI agents run code and processes, operate on files, access the network, and manage state.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/agent-sandbox-overview","quote":"Each sandbox runs in an independent VM environment that isolates compute, network, storage, and permissions across multiple dimensions.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/agent-sandbox-overview","quote":"Agent Sandbox supports multiple sandbox types, including code, browser, desktop, and mobile.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"Billing granularity: Resources are billed by the second. A usage period of less than 1 second is counted as 1 second.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000496/second (USD 0.017856/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000248/second (USD 0.008928/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.0000000703/GiB/second (USD 0.00025308/GiB/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000381/second (USD 0.013716/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000191/second (USD 0.006876/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000759/second (USD 0.027324/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000386/second (USD 0.013896/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.0000034/second (USD 0.01224/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"USD 0.00000167/second (USD 0.006012/hour)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"A free disk allowance of 15 GiB applies.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"Deep hibernation storage = Instance memory size + Instance disk size","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/billing-overview","quote":"Starting from 00:00 (UTC+8) on November 15, 2026, pre-warming fees are charged for custom templates.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/limits-of-agent-sandbox","quote":"Maximum runtime in the active state","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/limits-of-agent-sandbox","quote":"86400s","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/limits-of-agent-sandbox","quote":"300 instances","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/product-overview/open-service-area","quote":"The Flash Edition is available in the following regions:","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/pause-and-resume","quote":"Pause and resume are available only to allowlisted accounts.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/snapshots","quote":"The Snapshot feature currently supports only the second-generation runtime.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/snapshots","quote":"7 days from creation by default and expires automatically after that.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/runtime-gen1-vs-gen2","quote":"Based on lightweight virtual machine isolation, the second-generation runtime provides a runtime architecture for complex Agent workloads.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/runtime-gen1-vs-gen2","quote":"Supports only Ubuntu- or Debian-based images.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/network-access-control","quote":"Egress access control (outbound / egress)","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/build-sandbox-templates-with-ghcr-custom-images","quote":"builds Sandbox templates directly from images in a container registry.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/agent-sandbox/e2b-sdk-java-and-go","quote":"the Alibaba Cloud Function Compute team maintains the following E2B data-plane-compatible SDKs.","checked":"2026-11-15"},{"url":"https://www.alibabacloud.com/help/en/functioncompute/pay-as-you-go-of-fc-agent-sandbox","quote":"Billing rules are described for the Eco, Std, and Pro subscription plans.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-15","page":"https://aisandboxhunt.com/alibaba-agent-sandbox","score":{"code-sandbox":{"total":68,"rank":20,"of":63}},"comparablePrice":{"code-sandbox":12.570624},"benchmarks":[]},{"slug":"amp","name":"Amp","company":"Amp","website":"https://ampcode.com","docsUrl":"https://ampcode.com/docs/orbs","pricingUrl":"https://ampcode.com/pricing","categories":["coding-agent"],"tagline":"Coding agent that gives every thread its own cloud machine (an orb), with published per-size hourly prices.","summary":"Amp is a coding agent where each new thread can run in its own cloud machine, called an orb, with the repository cloned and tools installed. Orbs come in fixed sizes from 1 vCPU / 2 GB ($0.08/hour) to 16 vCPU / 44 GB, are billed by the minute while running, and pause automatically when idle with files and memory kept. The free Hobby tier pays the standard per-size price for hosted orbs; the $20 Megawatt and $200 Gigawatt tiers include a monthly block of orb time, and self-hosted orbs or your own runners cost nothing on these tiers. Model usage is separate: bring your own subscription or API key, or buy credits at provider list price.","isolation":"Separate sandboxed cloud machine per thread; Amp's security page says e2b provides the compute instances. The isolation technology is not stated by Amp.","isolationKind":"unknown","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["CLI","Web app","Desktop and mobile apps","Plugins (TypeScript)"],"regions":"Not stated for orbs. Amp's main infrastructure is on Google Cloud in the US.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":44},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per minute","billedWhile":"Only while the orb is running. An orb pauses 5 minutes after the agent stops working or 20 minutes after your last interaction, whichever is later; a paused orb costs nothing.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Hobby","monthlyFee":0,"includedUsageUsd":0,"limits":"No monthly fee and no included orb time: hosted orbs are pay-as-you-go at the standard per-size prices. Self-hosted orbs and your own runners are free and unlimited. Use your own model subscription or API keys, or buy credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Megawatt (Individual)","monthlyFee":20,"includedUsageUsd":0,"limits":"750 hours (45,000 minutes) of a1.small (2 vCPU / 4 GB) orbs a month; larger orbs use the hours faster. Also $20 of agent usage for Amp-served models and tools. Allowance resets monthly and does not roll over; after it runs out, buy credits or upgrade. a1.3xlarge is not available.","includedVcpuHours":1500,"includedGibHours":3000,"capped":false},{"name":"Gigawatt (Individual)","monthlyFee":200,"includedUsageUsd":0,"limits":"1,000 hours of a1.xxlarge (16 vCPU / 32 GB) orbs a month; larger orbs use the allowance faster and smaller ones more slowly. Also $200 of agent usage for Amp-served models and tools. Allowance resets monthly and does not roll over; after it runs out, buy credits.","includedVcpuHours":16000,"includedGibHours":32000,"capped":false}],"freeTierKind":"none","freeTier":"No free hosted orb time. The Hobby tier itself is free, and self-hosted orbs or your own runners cost nothing on Hobby, Individual and Teams tiers, but Amp-hosted orbs are pay-as-you-go.","notes":["The cloud machine is priced separately from the agent. Hosted orb sizes and standard prices: a1.tiny 1 vCPU / 2 GB $0.08/hour, a1.small 2 / 4 GB $0.17/hour, a1.medium 4 / 8 GB $0.33/hour, a1.large 8 / 16 GB $0.66/hour, a1.xxlarge 16 / 32 GB $1.32/hour, a1.3xlarge 16 / 44 GB $2.13/hour. Every size has a 60 GB disk.","The per-vCPU and per-GB rates we show come from Amp's published formula for the standard price (CPUs x $0.0504/hour + GB memory x $0.0162/hour, quoted for self-hosted orbs that match no size). It reproduces the hosted size prices to the cent, except a1.3xlarge, which Amp prices higher because of compute scarcity (the formula would give about $1.52/hour).","Ambiguity: Amp states memory in GB without saying GB or GiB; we use the GB rate as the GiB rate.","Paid-tier orb allowances are stated as hours of one size: 750 a1.small hours on Megawatt and 1,000 a1.xxlarge hours on Gigawatt. Larger orbs use the hours faster; the exact conversion between sizes is not stated on the pages checked. We convert them to vCPU-hours and GB-hours of the named size. At standard prices Megawatt's 750 a1.small hours are worth about $127.50 and Gigawatt's 1,000 a1.xxlarge hours about $1,320.","Paid tiers also include agent usage for Amp-served models and tools: $20 a month on Megawatt ($10 with the education discount) and $200 on Gigawatt. After that, credits are bought at list price.","Model usage is billed separately from orbs: use a ChatGPT, Claude or other subscription or your own API keys with no Amp token fees, or buy Amp credits charged at provider API prices with no markup for individuals and non-enterprise workspaces. Purchased credits expire 12 months after purchase.","There is no fixed run-time limit: the pause timeouts are idle timeouts, and an orb keeps running while the agent has work. A paused orb keeps its filesystem and memory, encrypted at rest, until the thread is deleted.","Each user can start a burst of 20 metered orbs, then one new orb every five minutes; later starts wait instead of failing.","Students and teachers can get Megawatt for $10 a month. Team workspaces have no extra fee; Enterprise pricing is custom. Self-hosted orbs are free on Hobby, Individual and Teams; Enterprise workspaces pay the standard price for them."]},"bestFor":["Running many agent tasks in parallel, each in its own paused-when-idle cloud machine.","Developers who want to use their own ChatGPT or Claude subscription or API keys for models.","Teams that want to run the same agent on their own machines at no extra cost."],"watchOut":["Hobby has no free hosted orb time; hosted orbs are billed at standard prices.","Paid-tier allowances are stated in hours of one orb size, and larger orbs use them faster.","The largest size (16 vCPU / 44 GB) is limited to the $200 tier and Enterprise and costs more per resource.","The isolation technology of hosted orbs is not documented."],"sources":[{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"a1.small | 2 | 4 GB | $0.17/hour","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"a1.3xlarge | 16 | 44 GB | $2.13/hour","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"Every current orb size includes 60 GB of disk.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"Orb usage is billed by the minute. A paused orb costs nothing.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"an orb pauses 5 minutes after the agent finishes or 20 minutes after your last interaction, whichever comes later","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"These are idle timeouts, not a limit on how long an orb can run.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"a1.3xlarge is available only to Gigawatt subscribers and Enterprise workspace members and is priced higher, due to the general scarcity of compute.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"CPUs × $0.0504/hour","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs/sizes-and-costs","quote":"GB memory × $0.0162/hour","checked":"2026-11-14"},{"url":"https://ampcode.com/pricing","quote":"45,000 minutes of orbin' time","checked":"2026-11-14"},{"url":"https://ampcode.com/pricing","quote":"$20 ∕ mo.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/markdown/pricing","quote":"750 hours (45,000 minutes) of a1.small orbs. Larger orbs use the hours faster.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/markdown/pricing","quote":"$20 of agent usage ($10 with the education discount) for Amp-served models and tools.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/markdown/pricing","quote":"| Gigawatt | $200/month | $200 of agent usage for Amp-served models and tools. After that, buy credits at list price. | 1,000 hours of a1.xxlarge orbs. |","checked":"2026-11-14"},{"url":"https://ampcode.com/pricing","quote":"Students and teachers can get Megawatt for $10 per month.","checked":"2026-11-14"},{"url":"https://ampcode.com/pricing","quote":"Pay-as-you-go for Amp-hosted orbs","checked":"2026-11-14"},{"url":"https://ampcode.com/pricing","quote":"An Amp paid tier gives you a big discount on orbs and includes Amp credits that you can use on any model.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/pricing","quote":"Included usage resets at the end of each monthly billing period and does not roll over.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/pricing","quote":"Orbs use a separate usage allowance.","checked":"2026-11-14"},{"url":"https://ampcode.com/docs/orbs","quote":"Each user can start a burst of 20 metered orbs.","checked":"2026-11-14"},{"url":"https://ampcode.com/security","quote":"e2b provides ephemeral compute instances for Amp orbs.","checked":"2026-11-14"},{"url":"https://ampcode.com/security","quote":"Amp saves a snapshot of a prepared orb for each project and orb size.","checked":"2026-11-14"},{"url":"https://ampcode.com/security","quote":"A paused orb’s filesystem and memory remain stored, encrypted at rest, until the thread is deleted.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/amp","score":{"coding-agent":{"total":53,"rank":7,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"anchor-browser","name":"Anchor Browser","company":"Anchor Browser","website":"https://anchorbrowser.io","docsUrl":"https://docs.anchorbrowser.io","pricingUrl":"https://anchorbrowser.io/pricing","categories":["browser"],"tagline":"Cloud Chromium for agents with managed logins, billed in credits: $0.05 per browser-hour plus $0.01 per browser.","summary":"Anchor Browser provides hosted Chromium sessions that agents drive over CDP, its SDKs or MCP, with managed authentication, proxies, captcha bypass and a built-in AI task agent. Usage is paid in credits worth $1 each: browser sessions you drive yourself cost $0.01 per browser created plus $0.05 per browser-hour, with proxy data and AI steps extra. Saved automation tasks are instead billed a flat 0.1 credits per successful run, or the infrastructure cost if higher. A free plan gives 5 credits every month.","isolation":"Not stated","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","REST API","MCP","Playwright","Puppeteer"],"regions":"US on Free and Starter; US, EU, Asia and Australia on Growth; any region on Enterprise.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Per minute (active time rounded to the nearest full minute), plus $0.01 per browser created","billedWhile":"While the browser session is active","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.05},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":5,"limits":"5 credits per month; up to 5 concurrent browsers; up to 500 browser sessions; 5 sessions created per minute; US only; no authenticated browsers, captcha bypass or geolocation. No overage listed.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Starter","monthlyFee":50,"includedUsageUsd":50,"limits":"50 credits per month, then $1.00 per credit; up to 25 concurrent browsers; 25 sessions created per minute; authenticated browsers, captcha bypass, any geolocation; US region.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":500,"includedUsageUsd":500,"limits":"500 credits per month, then $1.00 per credit; up to 50 concurrent browsers; Cloudflare verified browser agents; custom proxy; 30-day data retention.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Growth","monthlyFee":2000,"includedUsageUsd":2000,"limits":"2,000 credits per month, then $1.00 per credit; up to 200 concurrent browsers; 100 sessions created per minute; full stealth browser; US, EU, Asia and Australia regions.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan with 5 credits ($5 of usage) every month, up to 5 concurrent browsers.","notes":["One credit equals $1. Browser sessions driven directly are billed $0.01 per browser created plus $0.05 per browser-hour, rounded to the nearest full minute.","Ambiguity: the cost calculator on the pricing page uses $0.09 per browser-hour, while the pricing documentation lists $0.05. We show the documented $0.05 rate.","Anchor's proxy is $8/GB. Proxyless traffic or your own proxy is billed as egress at $0.20/GB.","AI steps cost $0.01 to $0.10 each depending on the model tier ($0.01 flat with your own model key).","Saved automation tasks are billed 0.1 credits ($0.10) per successful run, or the run's infrastructure cost if that is higher; failed runs pay only infrastructure.","Session length has no stated upper limit; the default maximum is 3 hours and can be raised. Batch sessions are limited to 24 hours.","The Team plan appears on the pricing page but not in the pricing documentation. Enterprise (500+ browsers, bring your own cloud or on-premises) is custom."]},"bestFor":["Agents that need to sign in to third-party web apps with managed authentication","Long-running browser sessions, since hourly browser time is low and there is no stated session cap","Repeated automation tasks that fit a flat per-run fee"],"watchOut":["Proxy traffic at $8/GB will usually cost more than browser time","Captcha bypass and geolocation need a paid plan; stealth needs Growth ($2,000/month)","Free and Starter plans run in the US only","The pricing page calculator and the docs show different hourly browser rates"],"sources":[{"url":"https://docs.anchorbrowser.io/pricing","quote":"$0.05 per browser hour","checked":"2026-11-13"},{"url":"https://docs.anchorbrowser.io/pricing","quote":"Charges are based on the total active time of the browser session, calculated to the nearest full minute.","checked":"2026-11-13"},{"url":"https://docs.anchorbrowser.io/pricing","quote":"$0.01 per browser","checked":"2026-11-13"},{"url":"https://docs.anchorbrowser.io/pricing","quote":"When you bring your own proxy or run proxyless, egress is billed at $0.2 per GB.","checked":"2026-11-13"},{"url":"https://docs.anchorbrowser.io/pricing","quote":"Every run of a task you created from a prompt or a demonstration costs 0.1 credits ($0.10)","checked":"2026-11-13"},{"url":"https://anchorbrowser.io/pricing","quote":"5 credits per month","checked":"2026-11-13"},{"url":"https://anchorbrowser.io/pricing","quote":"10574 sessions × (38 min ÷ 60) × $0.09/hour","checked":"2026-11-13"},{"url":"https://anchorbrowser.io/pricing","quote":"500 Credits per month","checked":"2026-11-13"},{"url":"https://docs.anchorbrowser.io/advanced/session-timeout","quote":"There is no upper limit on how long you can set the maximum duration.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/anchor-browser","score":{"browser":{"total":47,"rank":6,"of":12}},"comparablePrice":{"browser":0.05},"benchmarks":[]},{"slug":"anthropic-sandbox-runtime","name":"Anthropic sandbox-runtime","company":"Anthropic","website":"https://github.com/anthropics/sandbox-runtime","docsUrl":"https://github.com/anthropics/sandbox-runtime","pricingUrl":"https://github.com/anthropics/sandbox-runtime","categories":["code-sandbox"],"tagline":"Open-source CLI and library that wraps any process in OS-level filesystem and network restrictions, no container needed.","summary":"sandbox-runtime (srt) restricts what a process and its children can read, write and reach on the network, using the operating system's own sandboxing: Seatbelt on macOS, bubblewrap on Linux and a dedicated user account with a Windows Filtering Platform fence on Windows. Network traffic goes through local proxies that enforce domain allow and deny lists. It was built for Claude Code and is published as an npm package that works as a CLI or a library. It is a research preview under Apache-2.0.","isolation":"OS-level process sandbox (sandbox-exec/Seatbelt on macOS, bubblewrap on Linux, restricted user account on Windows) with proxy-based network filtering","isolationKind":"process","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/anthropics/sandbox-runtime","lastRelease":"2026-11-13","sdks":["CLI","TypeScript/JavaScript library"],"regions":"Runs locally on macOS, Linux or Windows.","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own machine","billedWhile":"Not applicable: no usage charges","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license.","notes":["Not a VM or container: the sandboxed process shares the host kernel, and isolation relies on OS sandbox features.","By default network access is denied and writes are allowed only to paths you list.","Published as a beta research preview; configuration formats may change."]},"bestFor":["Limiting what a local coding agent, MCP server or shell command can read, write and contact","Adding domain allow lists to processes on a developer machine without running containers","Embedding OS-level sandboxing in a TypeScript tool"],"watchOut":["Shares the host kernel; weaker than VM-based isolation for hostile code","Beta research preview; APIs and configuration may change","Behavior and limits differ between macOS, Linux and Windows"],"sources":[{"url":"https://github.com/anthropics/sandbox-runtime","quote":"A lightweight sandboxing tool for enforcing filesystem and network restrictions on arbitrary processes at the OS level, without requiring a container.","checked":"2026-11-13"},{"url":"https://github.com/anthropics/sandbox-runtime","quote":"srt uses native OS sandboxing primitives (sandbox-exec on macOS, bubblewrap on Linux) and proxy-based network filtering.","checked":"2026-11-13"},{"url":"https://github.com/anthropics/sandbox-runtime","quote":"By default, all network access is denied. You must explicitly allow domains.","checked":"2026-11-13"},{"url":"https://github.com/anthropics/sandbox-runtime","quote":"The Sandbox Runtime is a research preview developed for Claude Code to enable safer AI agents.","checked":"2026-11-13"},{"url":"https://github.com/anthropics/sandbox-runtime","quote":"npm install -g @anthropic-ai/sandbox-runtime","checked":"2026-11-13"},{"url":"https://github.com/anthropics/sandbox-runtime/releases/tag/v0.0.83","quote":"v0.0.83","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/anthropic-sandbox-runtime","score":{"code-sandbox":{"total":46,"rank":60,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"archil","name":"Archil","company":"Archil","website":"https://archil.com","docsUrl":"https://docs.archil.com/compute/sandboxes/introduction","pricingUrl":"https://archil.com/pricing","categories":["dev-box","code-sandbox"],"tagline":"Firecracker sandboxes whose root filesystem lives on a durable Archil disk, plus per-call serverless tool runs.","summary":"Archil sells durable disks for AI workloads and runs compute next to them: persistent sandboxes and one-shot serverless tool calls. Each sandbox is a Firecracker microVM whose root filesystem sits on an Archil disk, so installed packages survive stop, start and pause, and sandboxes can be forked. Sessions pause after at most 24 hours with memory intact and can be resumed. Sandbox CPU and memory are billed at $0.055 per vCPU-hour and $0.020 per GiB-hour on what is allocated, with paid plans set as monthly minimums that usage counts toward.","isolation":"One Firecracker microVM per sandbox with its own Linux kernel (KVM)","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","REST API","CLI"],"regions":"Disks in AWS us-east-1, us-west-2 and eu-west-1 and GCP us-central1; sandboxes run in Archil's public AWS regions. Other regions and bring-your-own-cloud on request or on Enterprise.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":32,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.055,"gibHour":0.02,"instanceHour":null,"instance":null,"storageGibMonth":0.2,"billingIncrement":"Not stated","billedWhile":"CPU and memory on what a running sandbox is allocated; stopped and paused sandboxes accrue no compute charges. Disk storage is billed separately for as long as data is kept.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Monthly allowance: 1 vCPU-hour and 2 GiB-hours of sandbox compute, 1,000 serverless tool calls, 5 disks of up to 10 GB each. Up to 5 concurrent sandboxes, sessions of at most 30 minutes, egress limited to a managed allowlist. Hard caps, no overage.","includedVcpuHours":1,"includedGibHours":2,"capped":true},{"name":"Developer","monthlyFee":10,"includedUsageUsd":10,"limits":"$10 monthly minimum that metered usage counts toward. 1,000 file systems, 20 concurrent sandboxes, external mounts, your own egress firewall rules and credential brokering, private sandbox images.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":500,"includedUsageUsd":500,"limits":"$500 monthly minimum that metered usage counts toward. 25,000 file systems, 250 concurrent sandboxes (500 more for $500 a month), 3 workspaces, SOC 2, shared Slack channel.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: each month 1 vCPU-hour and 2 GiB-hours of sandbox compute (an hour of the default 1 vCPU / 2 GiB sandbox), 1,000 tool calls and 5 disks of up to 10 GB; hard caps with no overage.","notes":["Paid plans are monthly minimums, not fees added on top: $6 of Developer usage bills as $10, $18 bills as $18.","A 1 vCPU / 2 GiB sandbox costs $0.095 an hour. Compute is billed on allocation, not on active use.","The storage rate we show is active storage at $0.20 per GiB-month. Archil-managed disks also pay archive storage of $0.025 per GB-month on all stored data; disks synced to your own bucket skip that charge.","Serverless tool calls cost $0.05 per 1,000 calls with execution included; each call runs in its own microVM and stops after 5 minutes.","Data transfer out of Archil costs $0.07 per GB; transfer within Archil is free.","A session pauses after at most 24 hours with processes intact and resumes as a new session; by default a sandbox also pauses after 30 minutes without a process connection.","Enterprise has custom pricing, unlimited concurrency and bring-your-own-cloud."]},"bestFor":["Agent workspaces whose files and installed packages must outlive any single VM","Sharing one data set between sandboxes, your own servers and S3 tools","Agent bash tools billed per call rather than per machine-hour"],"watchOut":["Compute is billed on allocated vCPU and memory while a sandbox runs, not on actual use","Sessions pause after 24 hours and must be resumed; Free sessions last at most 30 minutes","Free sandboxes can reach only a managed allowlist of hosts","Paid plans carry a monthly minimum of $10 or $500"],"sources":[{"url":"https://archil.com/pricing","quote":"CPU | $0.055 / vCPU-hour | Sandbox CPU usage counts toward your monthly minimum.","checked":"2026-11-13"},{"url":"https://archil.com/pricing","quote":"Memory | $0.020 / GiB-hour | Sandbox memory usage counts toward your monthly minimum.","checked":"2026-11-13"},{"url":"https://archil.com/pricing","quote":"Active storage | $0.20 / GiB-month","checked":"2026-11-13"},{"url":"https://archil.com/pricing","quote":"Free includes 5 disks of up to 10 GB each, 1,000 serverless tool calls, and 1 hour of sandbox compute a month (1 vCPU-hour and 2 GiB-hours, an hour of the default 1 vCPU, 2 GiB sandbox), with up to 5 concurrent sandboxes and sessions of up to 30 minutes. These are hard caps, with no overage.","checked":"2026-11-13"},{"url":"https://archil.com/pricing","quote":"Developer has a $10 monthly minimum and Team has a $500 monthly minimum.","checked":"2026-11-13"},{"url":"https://archil.com/pricing","quote":"A 1 vCPU, 2 GiB sandbox costs $0.095 an hour. Stopping or pausing a sandbox stops compute charges.","checked":"2026-11-13"},{"url":"https://docs.archil.com/compute/sandboxes/introduction","quote":"One Firecracker microVM per sandbox, with its own Linux kernel, isolated by hardware virtualization (KVM).","checked":"2026-11-13"},{"url":"https://docs.archil.com/compute/sandboxes/introduction","quote":"1 to 32 vCPU and 0.25 to 64 GiB of memory, on x86-64.","checked":"2026-11-13"},{"url":"https://docs.archil.com/compute/sandboxes/introduction","quote":"A session pauses after at most 24 hours. Resuming starts a new session.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/archil","score":{"dev-box":{"total":73,"rank":7,"of":38},"code-sandbox":{"total":74,"rank":9,"of":63}},"comparablePrice":{"dev-box":41.8,"code-sandbox":33.44},"benchmarks":[]},{"slug":"aws-agentcore-code-interpreter","name":"AWS Bedrock AgentCore Code Interpreter","company":"Amazon Web Services","website":"https://aws.amazon.com/bedrock/agentcore/","docsUrl":"https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/code-interpreter-tool.html","pricingUrl":"https://aws.amazon.com/bedrock/agentcore/pricing/","categories":["code-sandbox"],"tagline":"Managed AWS code-execution tool for agents, billed per second on CPU actually used and peak memory.","summary":"AgentCore Code Interpreter is a managed tool in Amazon Bedrock AgentCore that runs Python, JavaScript and TypeScript in isolated sessions for AI agents. CPU is billed only for actual consumption, so I/O wait costs nothing, while memory is billed on the peak used so far, both per second. Sessions default to 15 minutes and can be extended to 8 hours, with Sandbox or Public network modes and S3 access through an execution role.","isolation":"Isolated session per sandbox; AWS describes billing from microVM boot to microVM shutdown","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["AWS SDKs (including Python/boto3)","AWS CLI","AgentCore API"],"regions":"Not stated on the pages checked; available in AWS Regions where AgentCore is offered.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":true,"maxSessionHours":8,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.0895,"gibHour":0.00945,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per second, 1-second minimum; 128 MB minimum memory billing","billedWhile":"From microVM boot to session termination. CPU only for actual consumption (no charge during I/O wait); memory for the peak consumed up to each second, including idle periods.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"credit","freeTier":"No Code Interpreter free allowance. New AWS customers receive up to $200 in AWS Free Tier credits usable across AWS.","notes":["Rates are $0.0895 per vCPU-hour and $0.00945 per GB-hour; AWS writes GB, which we treat as GiB.","Billing includes system overhead in addition to the code's own resource use.","Network data transfer is charged separately at standard EC2 rates.","Default session timeout is 15 minutes; execution can be extended up to 8 hours.","Inline file upload is limited to 100 MB; files up to 5 GB can be moved through Amazon S3 from terminal commands.","Maximum vCPU and memory per session are not stated on the pages checked."]},"bestFor":["Agents already built on Amazon Bedrock or AgentCore that need a code tool","Workloads with long I/O waits, since idle CPU is not billed","Data analysis on files stored in Amazon S3"],"watchOut":["Only Python, JavaScript and TypeScript runtimes are provided","Memory is billed on peak use for the whole session, even while idle","Per-session CPU and memory limits are not published","Data transfer is billed separately at EC2 rates"],"sources":[{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"Code Interpreter |  | CPU | $0.0895 per vCPU-hour","checked":"2026-11-13"},{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"Memory | $0.00945 per GB-hour","checked":"2026-11-13"},{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"Billing is calculated per second, using actual CPU consumption and peak memory consumed up to that second, with a 1-second minimum","checked":"2026-11-13"},{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"which spans from microVM boot, initialization, active processing, idle periods, until session termination (microVM shutdown)","checked":"2026-11-13"},{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"Network data transfer charges apply at standard EC2 rates","checked":"2026-11-13"},{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"New AWS customers receive up to $200 in Free Tier credits.","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/code-interpreter-tool.html","quote":"The Code Interpreter tool provides support for a default execution time of 15 minutes, which can be extended for up to eight hours.","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/code-interpreter-tool.html","quote":"For inline upload, the file size can be up to 100 MB.","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/code-interpreter-resource-session-management.html","quote":"you can choose network settings (Sandbox or Public)","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/aws-agentcore-code-interpreter","score":{"code-sandbox":{"total":60,"rank":40,"of":63}},"comparablePrice":{"code-sandbox":16.104},"benchmarks":[]},{"slug":"aws-lambda-microvms","name":"AWS Lambda MicroVMs","company":"Amazon Web Services","website":"https://aws.amazon.com/lambda/lambda-microvms/","docsUrl":"https://docs.aws.amazon.com/lambda/latest/dg/lambda-microvms-guide.html","pricingUrl":"https://aws.amazon.com/lambda/pricing/","categories":["code-sandbox"],"tagline":"Firecracker microVMs from Lambda that suspend with memory and disk intact, billed per second on a baseline size.","summary":"Lambda MicroVMs launch Firecracker microVMs from a snapshot of your own Dockerfile-built image, each with its own HTTPS endpoint. You pick a baseline size (vCPU scales with memory at 1 vCPU per 2 GB) and the VM can burst to 4x that baseline, paying the baseline while running plus actual use above it, per second. Idle MicroVMs can be suspended with memory and disk preserved, paying only snapshot storage, and a MicroVM lives at most 8 hours.","isolation":"Firecracker microVM per MicroVM","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["AWS SDKs","AWS CLI","Lambda API"],"regions":"Multiple AWS Regions (account memory quotas are listed per Region); x86 is available in all MicroVMs Regions except a few in Asia Pacific and Europe.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":8,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.09969984,"gibHour":0.01320012,"instanceHour":null,"instance":null,"storageGibMonth":0.08,"billingIncrement":"per second","billedWhile":"While running: the configured baseline vCPU and memory for the whole time, plus actual use above the baseline when bursting. Suspended MicroVMs pay snapshot storage only; terminated ones pay nothing.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"credit","freeTier":"No MicroVMs free allowance is listed. New AWS customers receive up to $200 in AWS Free Tier credits usable across AWS.","notes":["Rates are the Arm (Graviton) rates in US East (N. Virginia) from AWS's own pricing examples: $0.0000276944 per vCPU-second and $0.0000036667 per GB-second, multiplied by 3,600. x86 rates were not captured. AWS writes GB, which we treat as GiB.","Sizes are fixed baselines from 0.5 GB / 0.25 vCPU to 8 GB / 4 vCPU (default 2 GB / 1 vCPU); each can burst to 4x, up to 32 GB / 16 vCPU. Disk is 8 to 32 GB depending on size.","Snapshot charges: $0.08/GB-month storage (images have a 1-week minimum retention), $0.0038/GB written on suspend and $0.00155/GB read on launch or resume.","Data transfer is charged at standard AWS rates.","Maximum lifetime per MicroVM is 8 hours, counting both running and suspended time.","Default account quota is 400 GB of configured memory per Region (1,024 GB in some large Regions)."]},"bestFor":["Interactive sandboxes that sit idle between user turns, since suspended VMs pay only storage","Per-tenant isolation with a dedicated HTTPS endpoint for each VM","Teams already on AWS that want VPC access from sandboxes"],"watchOut":["Hard 8-hour maximum lifetime per MicroVM, including suspended time","Every launch, suspend and resume adds per-GB snapshot read or write charges","Images must be built from a Dockerfile zip uploaded to S3 and snapshotted first","Baseline vCPU and memory are billed for the full running time, even when idle"],"sources":[{"url":"https://aws.amazon.com/lambda/pricing/","quote":"vCPU: 13.5M seconds × 1 vCPU × $0.0000276944/vCPU-second = $373.87","checked":"2026-11-13"},{"url":"https://aws.amazon.com/lambda/pricing/","quote":"Memory: 13.5M seconds × 2 GB memory × $0.0000036667/GB-second = $99.00","checked":"2026-11-13"},{"url":"https://aws.amazon.com/lambda/pricing/","quote":"All examples below are based on pricing in US East (N. Virginia), using ARM (Graviton) architecture.","checked":"2026-11-13"},{"url":"https://aws.amazon.com/lambda/pricing/","quote":"MicroVM image storage charges: 5 images × 2 GB × $0.08/GB-month = $0.80","checked":"2026-11-13"},{"url":"https://aws.amazon.com/lambda/pricing/","quote":"Snapshot write (suspend) charges: 12,000 cycles × 2 GB × $0.0038/GB = $91.20","checked":"2026-11-13"},{"url":"https://aws.amazon.com/lambda/pricing/","quote":"Snapshot read (resume) charges: 12,000 cycles × 2 GB × $0.00155/GB = $37.20","checked":"2026-11-13"},{"url":"https://aws.amazon.com/lambda/lambda-microvms/","quote":"Suspended MicroVMs incur no compute charges.","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/lambda/latest/dg/microvms-images.html","quote":"8 GB memory, 4 vCPU | 32 GB memory, 16 vCPU | 32 GB","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/lambda/latest/dg/microvms-images.html","quote":"You pay the baseline rate while your MicroVM is running and only pay for what you actively use above the baseline, billed per second.","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/lambda/latest/dg/gettingstarted-limits.html","quote":"Maximum execution duration per MicroVM | 8 hours (28,800 seconds) | No","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/lambda/latest/dg/lambda-microvms-guide.html","quote":"AWS Lambda MicroVMs deliver these core capabilities through Firecracker virtualization","checked":"2026-11-13"},{"url":"https://docs.aws.amazon.com/lambda/latest/dg/lambda-microvms-guide.html","quote":"Outbound access is configurable.","checked":"2026-11-13"},{"url":"https://aws.amazon.com/bedrock/agentcore/pricing/","quote":"New AWS customers receive up to $200 in Free Tier credits.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/aws-lambda-microvms","score":{"code-sandbox":{"total":65,"rank":29,"of":63}},"comparablePrice":{"code-sandbox":44.38722816},"benchmarks":[]},{"slug":"aws-nitro-sandboxes","name":"AWS Nitro Sandboxes","company":"Amazon Web Services","website":"https://aws.amazon.com/about-aws/whats-new/2026/11/nitro-sandboxes-eks-ecs-ec2/","docsUrl":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","pricingUrl":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","categories":["code-sandbox"],"tagline":"Hardware-isolated lightweight VMs on your own EC2 instances, one per EKS pod, ECS task or agent session.","summary":"Nitro Sandboxes is an EC2 feature that runs each workload in its own lightweight VM on the Nitro Hypervisor, next to a parent EC2 instance that lends it vCPUs and memory. On EKS a pod opts in with one line in its spec, and on ECS Managed Instances each task can run in its own sandbox; on plain EC2 you launch them yourself with Firecracker. Confidential mode hides sandbox memory from the parent instance and adds attestation that gates key release in AWS KMS, while direct mode trusts the parent. There is no extra charge: you pay for the EC2 instances (and any EKS Auto Mode or ECS Managed Instances fee).","isolation":"Lightweight VM per sandbox on the Nitro Hypervisor, launched through Firecracker on a parent EC2 instance; optional confidential mode with attestation","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Kubernetes API (kubectl/YAML) on EKS","Amazon ECS API","Firecracker on EC2"],"regions":"All commercial AWS Regions, on most Nitro-based EC2 instance types (not flex instances).","features":{"pauseResume":false,"snapshots":false,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: no charge for the sandboxes themselves; EC2 instances are billed at standard rates","billedWhile":"Not applicable: you pay for the parent EC2 instances while they run, however many sandboxes they host","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"none","freeTier":"The sandbox feature and its EKS and ECS integrations cost nothing extra, but there is no free compute: you pay standard EC2 prices for the parent instances.","notes":["No per-sandbox rate exists. Cost depends on the EC2 instance type you choose and how many sandboxes you pack onto it, so no vCPU or memory rate is listed.","EKS Auto Mode and ECS Managed Instances charge their standard management fee on top of EC2 if you use them.","Density: up to 5 sandboxes per parent vCPU and at most 896 per instance. A sandbox can have at most as many vCPUs as its parent, so the largest size depends on the instance type; no fixed maximum vCPU or memory is published. Each sandbox needs at least 256 MiB of memory.","On instances with two threads per core, a core runs only one sandbox at a time, so usable compute is about half the advertised vCPUs. Graviton instances are not affected.","Sandboxes live as long as the parent instance and you keep them; no maximum session length is published.","GPU passthrough is unavailable; on GPU instance types a sandbox runs CPU workloads only."]},"bestFor":["Teams on EKS or ECS that want a VM boundary per agent session or pod without a dedicated instance each","Running untrusted pull request code or LLM-generated code next to other tenants on shared instances","Confidential workloads that must stay hidden from the instance operator, with attestation-gated keys"],"watchOut":["Not a hosted API: you run and pay for the EC2 instances, clusters and orchestration yourself","No pause and resume, snapshot and restore, or live migration","No GPU passthrough","Without a block rule on the parent, a sandbox whose traffic is routed through the parent can reach the parent's instance metadata and its IAM role credentials"],"sources":[{"url":"https://aws.amazon.com/about-aws/whats-new/2026/11/nitro-sandboxes-eks-ecs-ec2/","quote":"AWS Nitro Sandboxes are available today on most Nitro-based EC2 instances, in all commercial regions worldwide, at no additional charge.","checked":"2026-11-15"},{"url":"https://aws.amazon.com/about-aws/whats-new/2026/11/nitro-sandboxes-eks-ecs-ec2/","quote":"you can safely execute LLM-generated code in an isolated environment of its own","checked":"2026-11-15"},{"url":"https://aws.amazon.com/blogs/compute/nitro-sandboxes-hardware-isolation-for-every-eks-pod-ecs-task-and-ec2-workload","quote":"You pay for the EC2 instances, plus the standard EKS Auto Mode or ECS Managed Instances management fee if you use those capabilities.","checked":"2026-11-15"},{"url":"https://aws.amazon.com/blogs/compute/nitro-sandboxes-hardware-isolation-for-every-eks-pod-ecs-task-and-ec2-workload","quote":"Network policies and security groups are enforced on the parent instance, outside the sandbox","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"There are no additional charges for using Nitro Sandboxes.","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"Currently, Firecracker is the only VMM that supports Nitro Sandboxes","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"You can run up to 5 Nitro Sandboxes for each vCPU of the parent instance, and at most 896 Nitro Sandboxes on one parent instance.","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"Give each Nitro Sandbox at least 256 MiB of memory.","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"Nitro Sandboxes do not support snapshot and restore, pause and resume, or live migration.","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"GPU passthrough into a Nitro Sandbox is currently unavailable.","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"Nitro Sandboxes are not available on flex instances","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/amazon-ec2-nitro-sandboxes.html","quote":"the Nitro Sandbox can reach the IMDS of the parent instance and read its AWS Identity and Access Management (IAM) role credentials","checked":"2026-11-15"},{"url":"https://docs.aws.amazon.com/eks/latest/userguide/sandboxing.html","quote":"Isolated and confidential pods are available in all commercial AWS Regions.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-15","page":"https://aisandboxhunt.com/aws-nitro-sandboxes","score":{"code-sandbox":{"total":45,"rank":62,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"azure-dynamic-sessions","name":"Azure Container Apps dynamic sessions","company":"Microsoft","website":"https://learn.microsoft.com/azure/container-apps/sessions","docsUrl":"https://learn.microsoft.com/en-us/azure/container-apps/sessions","pricingUrl":"https://azure.microsoft.com/en-us/pricing/details/container-apps/","categories":["code-sandbox"],"tagline":"Hyper-V isolated code interpreter sessions from prewarmed pools, billed per session-hour.","summary":"Dynamic sessions in Azure Container Apps hand out prewarmed sandboxes from a session pool, each isolated by a Hyper-V boundary. Built-in code interpreter pools run Python, Node.js or Shell and cost $0.03 per session-hour, billed in one-hour increments from allocation to deallocation. Custom container pools run your own image on Dedicated plan compute instead. Sessions end after an idle cooldown of 5 to 60 minutes.","isolation":"Hyper-V isolation per session","isolationKind":"gvisor","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["REST API","Azure CLI","LangChain, LlamaIndex, AutoGen and Semantic Kernel integrations","MCP server (Python pools)"],"regions":"38 Azure regions across the Americas, Europe, Asia Pacific, the Middle East and Africa.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.03,"instance":"One built-in code interpreter session (Python, Node.js or Shell); vCPU and memory per session not stated","storageGibMonth":null,"billingIncrement":"1 hour per allocated session","billedWhile":"From when a session is allocated until it is deallocated (after the idle cooldown or an explicit delete), rounded up to whole hours.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"credit","freeTier":"No dynamic sessions free allowance is listed. New Azure customers can get a $200 credit to use within 30 days.","notes":["Built-in code interpreter sessions are $0.03 per session-hour pay as you go; savings plans lower this to $0.026 (1 year) or $0.025 (3 years).","Billing is per allocated session in one-hour increments, so a 2-minute session is billed as a full hour.","Custom container session pools are billed on the Container Apps Dedicated plan for the E16 instances running the pool, not per session; those instance rates are not shown here.","Idle cooldown is 300 to 3,600 seconds; each code execution is limited to 220 seconds; file uploads are limited to 128 MB.","A pool can hold at most 600 concurrent sessions. Outbound network access is off by default and can be enabled per pool."]},"bestFor":["Running LLM-generated Python or JavaScript inside an existing Azure deployment","Agent frameworks with ready integrations such as LangChain or Semantic Kernel","Many short interactive sessions that need near-instant start from a prewarmed pool"],"watchOut":["Hourly billing increments make very short sessions relatively expensive","Each code execution stops after 220 seconds","Session vCPU and memory sizes for built-in interpreters are not published","Custom container pools bill whole Dedicated plan instances rather than per session"],"sources":[{"url":"https://azure.microsoft.com/en-us/pricing/details/container-apps/","quote":"Platform-managed built-in container (Python, Node.js, Shell) | $0.03 per session-hour","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/billing","quote":"For each allocated session, you're billed from the time it's allocated until it's deallocated in increments of one hour.","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/billing","quote":"Each custom container session pool runs on dedicated _E16_ compute instances.","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/sessions-code-interpreter","quote":"Each code interpreter session is fully isolated by a Hyper-V boundary and is designed to run untrusted code.","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/sessions-code-interpreter","quote":"Each code execution is limited to a maximum runtime of 220 seconds.","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/session-pool","quote":"The allowed range is between `300` and `3600` .","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/session-pool","quote":"Valid values are `EgressDisabled` (default) and `EgressEnabled` .","checked":"2026-11-13"},{"url":"https://learn.microsoft.com/en-us/azure/container-apps/sessions","quote":"Bring-your-own-container for custom workloads that require specific dependencies or runtime environments.","checked":"2026-11-13"},{"url":"https://azure.microsoft.com/en-us/pricing/purchase-options/azure-account","quote":"Get free services and a $200 credit to explore Azure for up to 30 days.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/azure-dynamic-sessions","score":{"code-sandbox":{"total":48,"rank":58,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"beam","name":"Beam","company":"Beam","website":"https://beam.cloud","docsUrl":"https://docs.beam.cloud/v2/sandbox/overview","pricingUrl":"https://www.beam.cloud/pricing","categories":["code-sandbox","gpu","dev-box"],"tagline":"Container sandboxes with optional GPUs, plus persistent VMs, billed per vCPU and GiB with three CPU price tiers.","summary":"Beam sandboxes are disposable containers for agent code, controlled from Python or TypeScript SDKs, with process, filesystem, networking and snapshot APIs; a sandbox can also run in a microVM, and GPUs can be attached. CPU and memory bill by the millisecond while a sandbox runs, at $0.05 per vCPU-hour and $0.01 per GiB-hour on the default Performance tier, with cheaper Shared and Spot tiers. Beam also sells persistent Linux VMs for agents at the same rates, GPU machines and serverless endpoints. Plans start at $20 a month with $20 of credit, and the engine behind Beam is open source (Beta9) and can be self-hosted.","isolation":"Isolated container by default; a sandbox can run in a microVM with its own kernel (use_vm=True). Persistent VMs are microVMs.","isolationKind":"container","openSource":true,"selfHost":true,"license":"AGPL-3.0","repoUrl":"https://github.com/beam-cloud/beta9","lastRelease":"2026-11-12","sdks":["Python","JavaScript/TypeScript","CLI","MCP server"],"regions":"Not stated for sandboxes. Bring-your-own-cloud runs in your AWS, GCP or Azure account.","features":{"pauseResume":null,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.05,"gibHour":0.01,"instanceHour":null,"instance":null,"storageGibMonth":0,"billingIncrement":"per millisecond","billedWhile":"Only while the sandbox is running, for its requested CPU and memory, until it is terminated or its TTL (default 600 seconds) expires.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"NVIDIA RTX 4090 24 GB (GPU only; CPU and RAM extra)","perHour":0.69},{"type":"NVIDIA RTX 5090 32 GB (GPU only; CPU and RAM extra; committed spend)","perHour":1.09}],"plans":[{"name":"Developer","monthlyFee":20,"includedUsageUsd":20,"limits":"$20 in credits each month. 30 concurrent vCPUs and 5 concurrent GPUs across VMs, sandboxes and serverless apps; 1 seat; 30-day log retention.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":89,"includedUsageUsd":0,"limits":"$89 a month plus usage. 1,000 concurrent vCPUs, 50 concurrent GPUs, 3 seats included ($25 per extra seat), private Slack channel.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free tier or trial credit is stated on the pricing page; the cheapest plan is $20 a month with $20 of credit.","notes":["Rates shown are the default Performance tier (no CPU sharing). Shared tier: $0.011 per vCPU-hour and $0.0055 per GiB-hour, up to 8 vCPU / 16 GiB. Spot tier: $0.0063 per vCPU-hour and $0.00135 per GiB-hour, preemptible.","GPU prices cover the GPU only. CPU and RAM attached to a GPU cost $0.18972 per vCPU-hour and $0.0198 per GiB-hour on every tier. On-demand GPU machines (H100 from $1.83/hour, B200 from $4.11/hour) are a separate product.","Storage is not billed yet (we show $0). Planned price: $0.0001 per GB-hour of data stored, with no free allowance.","No egress or bandwidth fees on any plan.","Persistent VMs bill at the same CPU-tier rates per second while running and can pause with memory intact; the smallest is 1 vCPU / 1 GiB with a 50 GiB disk at $0.06/hour. Sandboxes themselves have a TTL rather than pause/resume; state carries over through snapshots.","Sandbox defaults: 1 CPU, 128 MiB memory, 600-second lifetime; keep_warm_seconds=-1 keeps one until it is terminated. VM-backed sandboxes need at least 1 vCPU / 1 GiB.","Outbound access can be restricted by hostname, IP or CIDR.","Bring your own cloud: $0.019 per vCPU-hour and $0.009 per GB-hour management fee, with compute billed by your cloud provider. Volume discounts above $10K a month."]},"bestFor":["Agent sandboxes that sometimes need a GPU attached","Cost-sensitive batch work that can use the Spot or Shared CPU tiers","Teams that may later self-host the open-source engine"],"watchOut":["Sandboxes default to containers; microVM isolation must be requested","No free tier; the entry plan is $20 a month","The Developer plan caps concurrency at 30 vCPUs and 5 GPUs","Storage is free today but planned to be billed"],"sources":[{"url":"https://www.beam.cloud/pricing","quote":"$0.05/vCPU-hour + $0.01/GiB-hour","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing","quote":"$0.011/vCPU-hour + $0.0055/GiB-hour","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing","quote":"Pay by the millisecond, only when the sandbox is running.","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing","quote":"Sandboxes and serverless apps run in isolated containers, and a sandbox can run in a microVM with use_vm=True.","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing.md","quote":"| RTX 4090 PCIE | 24 GB | $0.69 | $0.000192 | Start instantly |","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing.md","quote":"| CPU | per vCPU | Attached to a GPU (same on every tier) | $0.18972 | $0.0000527 |","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing","quote":"$20/month $20 in credits each month","checked":"2026-11-13"},{"url":"https://www.beam.cloud/pricing","quote":"Storage is not billed yet. Planned pricing is based on data stored, with no free allowance.","checked":"2026-11-13"},{"url":"https://docs.beam.cloud/v2/sandbox/overview","quote":"The Python sandbox defaults to 1 CPU, 128 MiB of memory, Python 3.11, and a 600-second lifetime.","checked":"2026-11-13"},{"url":"https://docs.beam.cloud/v2/sandbox/overview","quote":"Use it to run untrusted or agent-generated code, fan out work from a snapshot, or attach a GPU to a short-lived task.","checked":"2026-11-13"},{"url":"https://github.com/beam-cloud/beta9","quote":"You can self-host Beta9 for free or choose managed cloud hosting through Beam.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/beam","score":{"code-sandbox":{"total":45,"rank":63,"of":63},"gpu":{"total":42,"rank":7,"of":7},"dev-box":{"total":44,"rank":36,"of":38}},"comparablePrice":{"code-sandbox":24.64,"gpu":null,"dev-box":30.8},"benchmarks":[]},{"slug":"bitrise-rde","name":"Bitrise Remote Dev Environments","company":"Bitrise","website":"https://bitrise.io/platform/remote-dev-environments","docsUrl":"https://docs.bitrise.io/en/bitrise-rde","pricingUrl":"https://bitrise.io/pricing","categories":["dev-box"],"tagline":"macOS and Linux cloud VMs on Bitrise CI infrastructure, billed per minute up to a monthly cost ceiling.","summary":"Bitrise Remote Dev Environments are on-demand macOS (Apple silicon) and Linux VMs running on the same stacks and caches as Bitrise CI, with iOS simulators and Android emulators available in a session. Sessions keep a persistent disk when terminated and can be restored later; they auto-terminate after five days of inactivity by default, configurable up to 30 days or off. Plans include minutes per operating system, then bill per minute up to a monthly ceiling, after which usage is free for the rest of the month. Agents connect through an MCP server, the CLI, an HTTP API or SSH.","isolation":"Full macOS or Linux VM per session","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["CLI","MCP server","HTTP API (experimental)","VS Code extension"],"regions":"Not stated on the pages checked; enterprise customers can get dedicated or private cloud environments.","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":false,"networkControls":false,"maxSessionHours":null,"maxVcpu":12,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.264,"instance":"Linux VM up to 4 vCPU / 16 GB, Basic plan overage rate ($0.0044 per minute)","storageGibMonth":null,"billingIncrement":"per minute","billedWhile":"While a session is running, after the plan's included minutes for that operating system; billing stops for the month once that operating system's monthly ceiling is reached. Terminated sessions keep their disk at no separate charge.","cpuActiveOnly":false,"unitVcpu":4,"unitGib":16,"browserHour":null},"gpu":[],"plans":[{"name":"Basic Tier 1","monthlyFee":20,"includedUsageUsd":2.112,"limits":"480 included minutes a month per OS (8 hours macOS and 8 hours Linux), 1 concurrent VM. macOS M2 Pro Large (6 vCPU / 14 GB), Linux up to 4 vCPU / 16 GB. Overage macOS $0.0288/min, Linux $0.0044/min; monthly ceiling macOS $99, Linux $39. $18/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Basic Tier 2","monthlyFee":40,"includedUsageUsd":4.224,"limits":"960 included minutes a month per OS, 3 concurrent VMs. Same machines and overage rates as Basic Tier 1; monthly ceiling macOS $297, Linux $117. $36/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Basic Tier 3","monthlyFee":60,"includedUsageUsd":6.336,"limits":"1,440 included minutes a month per OS, 5 concurrent VMs. Same machines and overage rates as Basic Tier 1; monthly ceiling macOS $495, Linux $195. $54/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro Tier 1","monthlyFee":40,"includedUsageUsd":4.224,"limits":"480 included minutes a month per OS, 1 concurrent VM. macOS M2 Pro X Large (12 vCPU / 28 GB), Linux up to 8 vCPU / 32 GB. Overage macOS $0.0576/min, Linux $0.0088/min; monthly ceiling macOS $199, Linux $79. $36/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro Tier 2","monthlyFee":80,"includedUsageUsd":8.448,"limits":"960 included minutes a month per OS, 3 concurrent VMs. Same machines and overage rates as Pro Tier 1; monthly ceiling macOS $597, Linux $237. $72/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro Tier 3","monthlyFee":120,"includedUsageUsd":12.672,"limits":"1,440 included minutes a month per OS, 5 concurrent VMs. Same machines and overage rates as Pro Tier 1; monthly ceiling macOS $995, Linux $395. $108/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"14-day free trial with Tier 1 allocation (8 hours per OS, 1 concurrent VM). A credit card is required at signup for verification but is not charged, the trial does not convert to a paid plan automatically, and trial sessions are limited to eight hours.","notes":["The rate shown is the Basic Linux overage rate: $0.0044 per minute = $0.264 per hour for a Linux VM up to 4 vCPU / 16 GB. Pro bills Linux at $0.0088 per minute ($0.528 per hour) on machines up to 8 vCPU / 32 GB, which matches two 4 vCPU / 16 GB units. The pricing pages do not say whether Pro charges the higher rate for a smaller Linux machine.","macOS overage is $0.0288 per minute ($1.728 per hour) on Basic (M2 Pro Large, 6 vCPU / 14 GB) and $0.0576 per minute ($3.456 per hour) on Pro (M2 Pro X Large, 12 vCPU / 28 GB).","Fees shown are for monthly billing; annual billing is 10% lower ($18 to $108 per month).","Each plan includes minutes separately for macOS and Linux. We value the included amount at the Linux overage rate only (for example Basic Tier 1: 480 minutes x $0.0044 = $2.112); the matching macOS minutes are worth 480 x $0.0288 = $13.824 on Basic Tier 1 but are not counted here.","Monthly ceiling: once overage for an operating system reaches its cap ($39 to $395 for Linux, $99 to $995 for macOS depending on plan and tier), that operating system is free for the rest of the month. Estimates that ignore the ceiling will overstate heavy use.","Session storage is not billed separately today. Terminated sessions keep their persistent disk indefinitely until deleted, unless the stack they were built from is removed; disks are not backed up.","Sessions auto-terminate after five days of inactivity by default (configurable up to 30 days, or off), and Bitrise may drain long-running machines for maintenance. Free-trial sessions are limited to eight hours.","Largest self-serve machines: macOS 12 vCPU / 28 GB and Linux 8 vCPU / 32 GB (Pro). The docs mention Linux up to 48 vCPUs and M4 machines, but those depend on the workspace and Enterprise contracts. Memory is stated in GB; we treat it as GiB.","Network filtering (default-deny egress) and custom VM images are listed as coming soon. Docker runs on Linux sessions only.","Agents run on your own AI subscription or API key. Remote Dev Environments are billed separately from Bitrise CI build credits. Enterprise pricing is based on committed capacity through sales."]},"bestFor":["Mobile teams whose agents need Xcode on real macOS, iOS simulators or Android emulators.","Persistent per-task dev machines that can be terminated and restored later with their disk.","Teams already on Bitrise CI that want agents to work on the same stacks and caches as their builds."],"watchOut":["Plans include only 8 to 24 hours a month per operating system; usage beyond that is billed per minute until the monthly ceiling.","Machine sizes are fixed per plan: Basic tops out at a 4 vCPU Linux VM and a 6 vCPU Mac.","Egress filtering and custom VM images are not available yet.","Session disks are not backed up and are lost if the underlying stack is removed."],"sources":[{"url":"https://bitrise.io/pricing","quote":"| Overage rate | macOS $0.0288/min · Linux $0.0044/min | macOS $0.0576/min · Linux $0.0088/min |","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"| Base fee billed annually | $18/month or $20 per month when billed monthly | $36/month or $40 per month when billed monthly | $54/month or $60 per month when billed monthly |","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"| Base fee billed annually | $36/month or $40 per month when billed monthly | $72/month or $80 per month when billed monthly | $108/month or $120 per month when billed monthly |","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"| Included minutes / month (per OS) | 480 minutes / month (8 hrs) | 960 minutes / month (16 hrs) | 1,440 minutes / month (24 hrs) |","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Once your bill for a VM (OS) type reaches its cap, that VM type is free for the rest of the month.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Two weeks, Tier 1 allocation (8 hours per OS, 1 concurrent VM), credit card required at signup for verification only.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"| Linux machine | Up to Medium (4 vCPU / 16GB) | Up to Large (8 vCPU / 32GB) |","checked":"2026-11-14"},{"url":"https://bitrise.io/platform/remote-dev-environments","quote":"A full macOS or Linux VM, isolated per session","checked":"2026-11-14"},{"url":"https://bitrise.io/platform/remote-dev-environments","quote":"Default-deny egress for every session with deep packet inspection (DPI).","checked":"2026-11-14"},{"url":"https://docs.bitrise.io/en/bitrise-rde/getting-started/rde-faq","quote":"Every session has an auto-terminate timer, defaulting to five days of inactivity. You can set it to anything up to 30 days, or turn it off so the session never auto-terminates.","checked":"2026-11-14"},{"url":"https://docs.bitrise.io/en/bitrise-rde/getting-started/rde-faq","quote":"Session storage isn't billed separately today.","checked":"2026-11-14"},{"url":"https://docs.bitrise.io/en/bitrise-rde/getting-started/rde-faq","quote":"There is no inactivity-based deletion of terminated sessions.","checked":"2026-11-14"},{"url":"https://docs.bitrise.io/en/bitrise-rde/getting-started/rde-faq","quote":"Sessions in free-trial workspaces are capped at eight hours","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/bitrise-rde","score":{"dev-box":{"total":53,"rank":30,"of":38}},"comparablePrice":{"dev-box":75.968},"benchmarks":[]},{"slug":"bitrise","name":"Bitrise CI","company":"Bitrise","website":"https://bitrise.io","docsUrl":"https://docs.bitrise.io","pricingUrl":"https://bitrise.io/pricing","categories":["ci-runner"],"tagline":"Mobile-focused CI/CD on managed macOS and Ubuntu VMs, with a free tier and plans priced by builds per month.","summary":"Bitrise is a CI/CD platform built for iOS and Android teams, running builds on managed macOS (Apple silicon) and Ubuntu virtual machines with a library of ready-made workflow steps. The free Hobby plan includes 500 build minutes a month; paid Starter and Pro plans are priced per month by concurrency and builds rather than by minute, and each machine type has a published per-minute cost used for plan minutes and pay-as-you-go. The smallest Linux machine, Linux 2S (2 vCPU, 8 GB), costs $0.0022 a minute and is available on Pro.","isolation":"Managed Ubuntu or macOS virtual machine per build","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Workflow YAML","MCP server"],"regions":"Not stated on the pricing page; Enterprise can use dedicated environments or self-hosted runners on AWS.","features":{"pauseResume":false,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":3.5,"maxVcpu":8,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.132,"instance":"Linux 2S: Ubuntu VM, 2 vCPU, 8 GB ($0.0022 per minute; Pro plan)","storageGibMonth":null,"billingIncrement":"per minute","billedWhile":"While a build runs; minutes count against the plan's allowance, then pay-as-you-go where the plan allows it.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Hobby","monthlyFee":0,"includedUsageUsd":0,"limits":"500 build minutes a month on Linux or macOS (Linux Medium 4 vCPU / 16 GB or M2 Pro Medium), 1 private app, 90-minute build timeout. Optional pay-as-you-go at 1.5x the per-minute rate, capped at $15 a month.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Starter","monthlyFee":99,"includedUsageUsd":0,"limits":"Unlimited usage with 1 concurrent build on Tier 1 ($99/month; $199 for 2 and $299 for 4 concurrent builds). Up to 3 private apps, Linux M (4 vCPU / 16 GB) and macOS Medium machines, 210-minute build timeout. $89/month billed annually.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Pro","monthlyFee":218,"includedUsageUsd":0,"limits":"From $218/month ($200 billed annually) for 250 builds a month; tiers include 3,000 to 24,000 zero-margin infrastructure minutes. Up to 10 concurrent macOS and 30 Linux builds; Linux 2S, Linux L (8 vCPU / 32 GB) and M2 Pro Large machines. Pay-as-you-go and extra build capacity available.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Hobby plan: free forever with 500 build minutes a month on Linux or macOS. New sign-ups also get a 30-day trial of the paid plans.","notes":["Rate shown is the published per-minute cost of Linux 2S (2 vCPU, 8 GB): $0.0022 x 60 = $0.132 per hour. Bitrise lists it for Pro and the Pro trial; Hobby and Starter only offer 4 vCPU Linux machines at $0.0044 per minute ($0.264 per hour).","Paid plans are priced mainly by builds per month and concurrency, not by minute. Starter has unlimited minutes within its concurrency limit; Pro tiers include a monthly pool of zero-margin infrastructure minutes that machines draw from at their per-minute cost. Pro tier prices above the first tier are not listed on the pricing page.","Pay-as-you-go: on Hobby each extra minute costs 1.5x its rate, up to $15 a month; on Pro extra minutes and up to 50% more builds are charged at a premium rate that is not published. Pay-as-you-go charges under $5 are not billed.","Hobby's 500 minutes are a minute allowance rather than a dollar credit; on Linux Medium they would be worth $2.20 at list rates.","Larger Linux machines (up to 48 vCPU / 192 GB) and M4 Pro Macs are on Enterprise or Bitrise's other plan families; the largest self-serve Linux machine is Linux L (8 vCPU / 32 GB). Memory is stated in GB; we treat it as GiB.","Build timeout is 90 minutes on Hobby, 210 minutes on Starter and Pro, and up to 4 hours on Enterprise.","Bitrise also sells Build Hub, which runs GitHub Actions jobs on Bitrise machines from $15 a month (Linux 2 vCPU / 8 GB at $0.00625 per minute from a monthly budget), and Remote Dev Environments, listed separately."]},"bestFor":["iOS and Android teams that need managed macOS and Xcode builds","Small mobile projects that fit in 500 free minutes a month","Teams that prefer a fixed monthly price per concurrent build over per-minute billing"],"watchOut":["Paid plans are priced by builds and concurrency, so per-minute costs are hard to compare","The cheapest 2 vCPU Linux machine needs the Pro plan, which starts at $218 a month","Builds time out after 210 minutes on self-serve plans (90 on Hobby)","Pro pay-as-you-go rates are not published"],"sources":[{"url":"https://bitrise.io/pricing","quote":"Linux 2S | Ubuntu VM, 2 vCPU, 8 GB RAM | Mobile DevOps Platform Plan, Pro and Pro trial | $0.0022","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Linux Medium | Ubuntu VM, 4vCPU, 16GB RAM @3.1GHz | Hobby | $0.0044","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"The Hobby plan is free, with 500 build minutes a month on Linux or macOS, 1 private app, and unlimited public apps.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Each minute is charged at 1.5x per minute and counts toward your $15 cap.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"From $89 per month when billed annually, or $99 per month when billed monthly","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"From $200 per month when billed annually, or $218 per month when billed monthly","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Use the Bitrise agent to run your builds on your AWS or on-premise infrastructure.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Perfect for growing teams who need increased collaboration. Starts from 250 builds/month.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"Zero Margin Infra minutes/month","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"With the Starter plan, you can get your new pipeline up and running with unlimited build minutes.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"On our Hobby plan, your build times out at 90 minutes, while on Starter and Pro, it can run up to 210 minutes.","checked":"2026-11-14"},{"url":"https://bitrise.io/pricing","quote":"If your PAYG charges are less than $5, you won’t be billed.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/bitrise","score":{"ci-runner":{"total":74,"rank":3,"of":10}},"comparablePrice":{"ci-runner":13.200000000000001},"benchmarks":[]},{"slug":"blacksmith","name":"Blacksmith","company":"Blacksmith Software Inc","website":"https://www.blacksmith.sh","docsUrl":"https://docs.blacksmith.sh/introduction/why-blacksmith","pricingUrl":"https://www.blacksmith.sh/pricing","categories":["ci-runner"],"tagline":"GitHub Actions runners in Firecracker microVMs on gaming-class CPUs, $0.004 per minute for 2 vCPU x64.","summary":"Blacksmith runs GitHub Actions jobs in ephemeral Firecracker microVMs on its own bare-metal fleet in the US and EU, switched in by changing the runs-on label. Linux x64 and ARM, Windows and macOS M4 runners are available from 2 to 32 vCPU, with a co-located cache, Docker layer caching and sticky disks. Ubuntu x64 costs $0.004 per minute for 2 vCPU and scales linearly with size. Every organization gets 3,000 free minutes a month, and there is no concurrency cap.","isolation":"Ephemeral Firecracker microVM per job on Blacksmith-run bare metal","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["GitHub Actions runs-on labels","CLI"],"regions":"us-east (Ashburn), us-west (Phoenix), eu-west (Amsterdam) and eu-central (Frankfurt); moving an organization to another region is done through support.","features":{"pauseResume":false,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":128},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.24,"instance":"Ubuntu x64 runner: 2 vCPU, 8 GB, Linux ($0.004 per minute)","storageGibMonth":0.5,"billingIncrement":"per minute","billedWhile":"While a job runs on the runner; the microVM is destroyed when the job ends. Sticky disks and Docker layer cache are billed per GB-month.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"3,000 free minutes a month, then per-minute rates; no concurrency limit. Add-ons: Docker layer caching and sticky disks $0.50/GB-month, static IPs $100/IP-month.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"3,000 free minutes every month; no credit card required to start. The pricing page does not say which runner size the free minutes are counted on.","notes":["Rate shown is Ubuntu x64 2 vCPU / 8 GB: $0.004 per minute x 60 = $0.24 per hour. x64 prices scale linearly: 4 vCPU $0.008, 8 vCPU $0.016, 16 vCPU $0.032, 32 vCPU $0.064 per minute.","Ubuntu ARM is cheaper at $0.0025 per minute for 2 vCPU / 6 GB ($0.15 per hour). Windows x64 is $0.008 per minute for 2 vCPU and macOS M4 $0.08 per minute for 6 vCPU / 24 GB.","The 3,000 free minutes are not converted to a dollar value here because the pricing page does not say whether they apply to all runner sizes; on the 2 vCPU x64 runner they would be worth $12.","Storage rate shown is for sticky disks and Docker layer caching ($0.50 per GB-month). The dependency cache includes 25 GB free per repository each week.","Jobs may get a free upgrade to the next runner size when the fleet has spare capacity.","Memory is stated in GB; we treat it as GiB. Startup discounts and an open-source program are offered on request; Enterprise adds a 99.9% SLA."]},"bestFor":["Teams that want faster single-thread CPUs for GitHub Actions with a one-line change","Docker-heavy pipelines that benefit from layer caching and sticky disks","Bursty workloads that shard tests widely, since concurrency is not capped"],"watchOut":["Caching add-ons are billed per GB-month on top of minutes","Static IPs cost $100 per IP per month","macOS runners cost 20 times the 2 vCPU Linux rate per minute","Only works with GitHub Actions"],"sources":[{"url":"https://www.blacksmith.sh/pricing","quote":"Ubuntu x64 $0.004 /min","checked":"2026-11-13"},{"url":"https://www.blacksmith.sh/pricing","quote":"3,000 free min/mo.","checked":"2026-11-13"},{"url":"https://www.blacksmith.sh/pricing","quote":"Sticky disks (+$0.50/GB/mo.)","checked":"2026-11-13"},{"url":"https://www.blacksmith.sh/pricing","quote":"Static IPs (+$100/IP/mo.)","checked":"2026-11-13"},{"url":"https://www.blacksmith.sh/pricing","quote":"Ubuntu ARM $0.0025 /min","checked":"2026-11-13"},{"url":"https://docs.blacksmith.sh/blacksmith-runners/overview","quote":"Blacksmith runs Linux and Windows jobs in ephemeral Firecracker microVMs that boot in under 3 seconds.","checked":"2026-11-13"},{"url":"https://docs.blacksmith.sh/blacksmith-runners/overview","quote":"Blacksmith is available in us-east (Ashburn), us-west (Phoenix), eu-west (Amsterdam), and eu-central (Frankfurt).","checked":"2026-11-13"},{"url":"https://docs.blacksmith.sh/blacksmith-runners/overview","quote":"blacksmith-32vcpu-ubuntu-2404 | 32 vCPU | 128 GB | 1.5 TB","checked":"2026-11-13"},{"url":"https://www.blacksmith.sh/security","quote":"We run jobs in a secure, isolated Firecracker microVM on our own bare-metal hardware, protected by a private VPN.","checked":"2026-11-13"},{"url":"https://docs.blacksmith.sh/introduction/quickstart","quote":"Plus, enjoy 25GB of free storage per repository each week","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/blacksmith","score":{"ci-runner":{"total":62,"rank":8,"of":10}},"comparablePrice":{"ci-runner":24},"benchmarks":[]},{"slug":"blaxel","name":"Blaxel","company":"Blaxel","website":"https://blaxel.ai","docsUrl":"https://docs.blaxel.ai/Sandboxes/Overview","pricingUrl":"https://blaxel.ai/pricing","categories":["code-sandbox"],"tagline":"MicroVM sandboxes that go to standby when idle and resume with memory and processes intact.","summary":"Blaxel Sandboxes are microVMs for running agent and LLM-generated code, with a REST API, an MCP server and SDKs for files and processes. A sandbox without an active connection moves to standby after a few seconds; its memory, filesystem and running processes are snapshotted and resume in about 25 ms. Compute is billed per GB of allocated memory per second only while the sandbox is active; CPU is allocated from the memory size and not billed separately. Standby sandboxes pay only snapshot storage.","isolation":"microVM (hardware-isolated, one per sandbox; hypervisor not named)","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","Go","CLI","REST API","MCP server"],"regions":"us-pdx-1 (Oregon), us-was-1 (North Virginia), eu-lon-1 (London), eu-fra-1 (Frankfurt); not all regions are self-service by default","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":0.0414,"instanceHour":null,"instance":null,"storageGibMonth":0.2,"billingIncrement":"per second (rate published per GB of RAM per second)","billedWhile":"allocated memory while the sandbox is active (has an active connection); standby time is not billed for compute, only snapshot storage","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pay-as-you-go (Tier 0)","monthlyFee":0,"includedUsageUsd":0,"limits":"Up to 10 sandboxes; sandbox TTL capped at 7 days; free credits of up to $200 (one-time, not monthly); community support","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pay-as-you-go (Tier 1)","monthlyFee":20,"includedUsageUsd":20,"limits":"Requires $20 of top-ups in the trailing 30 days; the top-up is prepaid credit for usage, not a fee. Up to 50 sandboxes; sandbox TTL capped at 30 days","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pay-as-you-go (Tier 2)","monthlyFee":50,"includedUsageUsd":50,"limits":"Requires $50 of top-ups in the trailing 30 days (prepaid credit for usage). Up to 200 sandboxes; higher tiers (up to Tier 9, 100,000+ sandboxes) are listed in the console","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Up to $200 in free credits at sign-up; Tier 0 quotas (10 sandboxes, 7-day maximum sandbox lifetime). Promotional credits do not count toward tier upgrades.","notes":["Sandbox compute is $0.0000115 per GB of RAM per second = $0.0414 per GB-hour (x3600). The page says GB; memory is set in MB (e.g. 4096), so we treat it as the per-GiB rate. This is an ambiguity in the vendor's units.","CPU is not billed separately: it is allocated from memory (docs: 8 GB gives 4 cores; the overview page says 16 GB gives 6 cores while a troubleshooting page says 16 GB gives 8 cores). So we show no separate vCPU rate; the calculator charges memory only.","Only active time is billed for compute. A sandbox goes to standby about 15 seconds after its last connection closes (the pricing FAQ says about 5 seconds); idle WebSocket/TCP connections time out after 15 minutes. Keep-alive techniques keep it billing.","Standby snapshots (memory + filesystem) cost $0.20 per GB-month; sandbox images $0.045 per GB-month; volumes $0.12 per GB-month of provisioned size, billed for as long as the volume exists.","Internet egress and internal traffic are listed as included. Agent Drive, dedicated egress IPs and the egress proxy are free during beta.","Tiers are set by the 30-day top-up volume; if top-ups fall below the threshold the workspace is downgraded and resource creation may be blocked. Add-ons: email support $800/month + 3% of usage, dedicated support $1,600/month + 10% of usage, HIPAA BAA $250/month. A Custom plan (up to 256 GB RAM per sandbox, bring-your-own-metal) is quote-only.","Per-sandbox vCPU and memory maximums for self-serve tiers are not published on public pages (shown only in the console quotas page)."]},"bestFor":["Agents that sit idle between turns and must resume quickly with processes intact","Many long-lived per-user sandboxes kept in standby at storage-only cost","Workloads that need domain allowlists or secret injection on outbound traffic"],"watchOut":["Any open connection or keep-alive keeps the sandbox active and billing for its full memory allocation","Tier 0 and Tier 1 sandboxes are deleted after at most 7 and 30 days; longer persistence needs a higher top-up tier","Writable filesystem lives in RAM (about 50% of memory), so disk-heavy work needs more memory or a volume","Manual snapshots and forking are in private preview on the Carbon generation and not recommended for production"],"sources":[{"url":"https://blaxel.ai/pricing","quote":"$0.0000115 / GB RAM · second","checked":"2026-11-14"},{"url":"https://blaxel.ai/pricing","quote":"Up to $200 in free credits","checked":"2026-11-14"},{"url":"https://docs.blaxel.ai/Sandboxes/Overview","quote":"CPU resources are allocated accordingly by Blaxel based on your selected memory allocation and are not charged separately.","checked":"2026-11-14"},{"url":"https://docs.blaxel.ai/Sandboxes/Expiration","quote":"Some quota tiers enforce expiration dates (Tier 0 and 1, with respectively up to 7 and 30 days).","checked":"2026-11-14"},{"url":"https://docs.blaxel.ai/Infrastructure/Regions","quote":"Available deployment regions for Blaxel infrastructure including Oregon, Virginia, London, and Frankfurt","checked":"2026-11-14"},{"url":"https://docs.blaxel.ai/Sandboxes/Fork","quote":"This feature is currently in private preview and is not recommended for production use.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/blaxel","score":{"code-sandbox":{"total":63,"rank":31,"of":63}},"comparablePrice":{"code-sandbox":29.145599999999998},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":351,"unit":"ms","better":"lower","date":"2026-11-13","rank":7,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":60706,"unit":"ms","better":"lower","date":"2026-11-13","rank":12,"of":25},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":19.48,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":6,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":12.1,"unit":"s","better":"lower","date":"2026-09-29","rank":3,"of":11}]},{"slug":"boat","name":"boat","company":"boat","website":"https://boat.dev","docsUrl":"https://docs.boat.dev/","pricingUrl":"https://docs.boat.dev/pricing","categories":["dev-box","code-sandbox"],"tagline":"Full Ubuntu VMs with SSH, Docker and a desktop, sold in fixed sizes per hour from a $20 monthly plan.","summary":"boat runs each sandbox as its own Ubuntu virtual machine with root, SSH, Docker, a browser desktop and pre-installed coding agents, driven by the boat CLI, Python and TypeScript SDKs or a REST API. Sandboxes come in four fixed sizes billed per second while running, from $0.018 an hour for 2 vCPU / 4 GB to $0.200 for 16 vCPU / 32 GB. A stopped sandbox has no compute charge and keeps its disk for at least 14 days; memory and processes are not kept after a short window. Plans start at $20 a month, and the plan price comes back as VM time.","isolation":"Virtual machine per sandbox with its own kernel, disk and IP address","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI","REST API"],"regions":"Germany, Finland and France; no US or Asia region. Snapshot copies can be stored outside the EU.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":null,"networkControls":false,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.018,"instance":"small: 2 vCPU, 4 GB RAM, 51 GB disk. Other sizes: default 4 vCPU / 8 GB / 51 GB at $0.036/h, large 8 vCPU / 16 GB / 111 GB at $0.072/h, xlarge 16 vCPU / 32 GB / 251 GB at $0.200/h ($100 plan or higher).","storageGibMonth":null,"billingIncrement":"per second","billedWhile":"Wall-clock while the sandbox runs, at the size's hourly price. Nothing for compute while stopped.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":4,"browserHour":null},"gpu":[],"plans":[{"name":"$20/mo","monthlyFee":20,"includedUsageUsd":20,"limits":"$20 of VM time per month (555 hours of the default size), not carried over. 100 sandboxes at once; 12 starts per minute, 60 per hour, 200 per day included; 10 saved snapshots. Sizes up to large. Overage paid from a prepaid wallet.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"$100/mo","monthlyFee":100,"includedUsageUsd":100,"limits":"$100 of VM time per month (2,777 hours of the default size). 300 sandboxes at once; 30 / 210 / 840 starts per minute / hour / day included; 50 saved snapshots. Unlocks xlarge.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"$500/mo","monthlyFee":500,"includedUsageUsd":500,"limits":"$500 of VM time per month (13,888 hours of the default size). 1,000 sandboxes at once; 65 / 420 / 1,680 starts per minute / hour / day included; 250 saved snapshots.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"$2000/mo","monthlyFee":2000,"includedUsageUsd":2000,"limits":"$2,000 of VM time per month (55,555 hours of the default size). 2,000 sandboxes at once; 90 / 600 / 2,400 starts per minute / hour / day included; 1,000 saved snapshots.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"7-day free trial on the first plan, card required: 25 compute hours in total, 2 sandboxes at once, small and default sizes only, auto-stop of 2 hours at most, no saved snapshots. There is no free tier without a card.","notes":["boat sells fixed sizes, not per-resource rates. We show the small size ($0.018/h for 2 vCPU / 4 GB) as the unit; small, default and large scale linearly from it. xlarge (16 vCPU / 32 GB) is $0.200/h, more than 8 small units ($0.144), so estimates at that size from the unit price are low.","RAM is stated in GB without saying GB or GiB; we treat the GB figures as GiB.","A plan is a monthly price that returns as VM time at the hourly rates; included time expires at the end of the billing period. Usage past it is paid from a prepaid wallet at the same rates. Organizations pay the plan price once per seat.","Starts (create, fork, resume) above the included number cost $0.03 to $0.07 each; above 6 times the hourly or daily allowance (3 times per minute) a start fails. Saved snapshots past the included count cost $0.056 per snapshot per day. The AI credentials proxy costs $1 per day it is on plus $1 per 1,000 requests.","Network egress is not billed. Disk is included in each size; there is no per-GB storage rate.","Stop keeps the disk only; boat keeps the machine for 10 minutes (not billed), and a resume in that window usually keeps memory and processes. A stopped sandbox's disk is kept for at least 14 days, after which retention cleanup can delete it once enabled (currently report-only). Saved snapshots have no scheduled expiry.","Auto-stop defaults to 1 hour after create or resume; it can be set up to 30 days or turned off on a paid plan.","boat has no API setting for network rules; outbound traffic is open and can be limited only with firewall rules inside the sandbox, which the root user can remove.","Pages differ on networking: the homepage says sandboxes have a public IPv6 address and outbound IPv4, while the docs index says each has a public IPv4; the limits text says some sandboxes have no IPv4 address of their own.","Custom base images are not described; saved snapshots of a prepared sandbox are used as templates instead."]},"bestFor":["Full Linux dev machines for agents, with root, Docker, systemd, SSH and a desktop.","Steady, CPU-heavy workloads where a flat hourly price per size is easier to predict than active-CPU billing.","Teams in Europe that want sandboxes hosted in Germany, Finland and France."],"watchOut":["A $20 monthly plan is the minimum; the trial needs a card.","No US or Asia region and no GPU on any size.","Stopping keeps only the disk; memory and running processes are lost after about 10 minutes.","Starts above the plan's daily allowance cost extra and can fail; nothing is queued."],"sources":[{"url":"https://docs.boat.dev/llms.txt","quote":"`small`: 2 vCPU, 4 GB RAM, 51 GB disk, $0.018/h","checked":"2026-11-14"},{"url":"https://docs.boat.dev/llms.txt","quote":"`xlarge`: 16 vCPU, 32 GB RAM, 251 GB disk, $0.20/h. Needs the $100 plan or higher.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/pricing","quote":"One `default` sandbox that runs a full month (730 h) costs $26.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/pricing","quote":"You pay per second, only while a sandbox runs. A stopped sandbox costs no machine time.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/pricing","quote":"A plan is a monthly price that your card pays. It is not a fee. Every dollar comes back as VM time at the rates above.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/pricing","quote":"$0.056 per snapshot per day ($1.68 for 30 days)","checked":"2026-11-14"},{"url":"https://docs.boat.dev/pricing","quote":"Network traffic. boat does not bill egress.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/free-trial","quote":"Your first plan starts with a free 7-day trial. Checkout asks for a card. There is no free tier without a card.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/limits","quote":"Auto-stop | Default 1 hour. Maximum 30 days, or off. Trial: 2 hours maximum.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/security","quote":"Each sandbox is its own virtual machine, with its own kernel, disk and IP address.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/security","quote":"boat has no API setting for network rules.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/llms.txt","quote":"boat keeps the machine 10 minutes (not billed): a resume in that window usually gets the same machine with memory and processes alive.","checked":"2026-11-14"},{"url":"https://boat.dev","quote":"Sandboxes run on machines in Germany, Finland and France. There is no US or Asia region.","checked":"2026-11-14"},{"url":"https://docs.boat.dev/llms.txt","quote":"$20/month: 100 sandboxes at once, starts 12/min, 60/h, 200/day","checked":"2026-11-14"},{"url":"https://docs.boat.dev/llms.txt","quote":"$2000/month: 2,000 at once, starts 90/min, 600/h, 2,400/day","checked":"2026-11-14"},{"url":"https://docs.boat.dev/llms.txt","quote":"Free trial: 7 days, card required. Until the first payment: 25 h of compute in total, 2 sandboxes at once","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/boat","score":{"dev-box":{"total":73,"rank":8,"of":38},"code-sandbox":{"total":71,"rank":14,"of":63}},"comparablePrice":{"dev-box":20,"code-sandbox":20},"benchmarks":[{"source":"aisandboxbench","metric":"Cold start (median)","value":1.125,"unit":"s","better":"lower","date":"2026-10-23","rank":6,"of":9},{"source":"aisandboxbench","metric":"Agent tool-call replay (median)","value":43.256,"unit":"s","better":"lower","date":"2026-10-23","rank":9,"of":9},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":26.69,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":1,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":18.7,"unit":"s","better":"lower","date":"2026-09-29","rank":7,"of":11}]},{"slug":"boxd","name":"boxd","company":"boxd","website":"https://boxd.sh","docsUrl":"https://docs.boxd.sh/","pricingUrl":"https://boxd.sh/pricing/","categories":["dev-box"],"tagline":"Persistent KVM microVMs that fork, checkpoint and hibernate, billed on active vCPU, resident RAM and written disk.","summary":"boxd runs persistent Linux machines in microVMs on its own KVM-based VMM, hosted in Frankfurt and New York. Machines default to 2 vCPU, 8 GiB and a 100 GiB disk, can be forked while running, checkpointed and snapshotted, and hibernate after 4 hours without network traffic, keeping their processes. Billing is credit-based and metered per second on active vCPU, resident memory and disk written, not on provisioned size. New organizations get 30 in free credits when they add a payment method, and the platform can also be self-hosted, free up to 16 vCPUs.","isolation":"Dedicated microVM per machine on boxd's own Rust VMM built on KVM","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI","REST API","MCP server"],"regions":"Frankfurt (Germany) and New York; machines and data in Frankfurt stay in the EU.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.047,"gibHour":0.014,"instanceHour":null,"instance":null,"storageGibMonth":0.07008,"billingIncrement":"Metered every second, settled every 15 minutes, no minimum charge per machine","billedWhile":"vCPU only while running and only for active vCPU; RAM on resident memory while running or in standby; disk on space written in every state, including hibernation.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"Credit-based, no monthly or seat fee. Up to 100 machines per organization, machines up to 8 vCPU / 32 GiB. Auto top-up adds 20 at a time once credits run out (can be changed or turned off).","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"30 of free credits (EUR or USD, by the organization's currency) when a payment method is added; nothing is charged when the card is added.","notes":["Currency: boxd bills an organization in euros or US dollars, chosen when it is created. The rates shown are boxd's own USD prices: $0.047 per active vCPU-hour, $0.014 per GiB-hour of resident RAM and $0.000096 per GiB-hour of disk written. The euro rates are EUR 0.039, EUR 0.012 and EUR 0.00008.","Disk storage converted per 730-hour month: $0.000096 x 730 = $0.07008 per GiB-month (EUR 0.0584), charged on space written, not the 100 GiB provisioned.","Only active vCPU is billed: a 2 vCPU machine busy 10% of an hour pays for 0.2 vCPU-hours. boxd quotes a default 2 vCPU / 8 GiB machine at about EUR 0.17 / USD 0.21 per hour fully busy, and about EUR 0.03 / USD 0.04 per hour at 0.2 vCPU active, 2 GiB held and 20 GiB written.","States: running pays vCPU, RAM and disk; standby (paused, memory kept warm, resumes in under 1 ms) pays RAM and disk; hibernated (memory written to disk, wakes in about 156 ms) pays disk only. A machine hibernates on its own after 4 hours with no network traffic by default.","Self-serve sizes per the FAQ are 1 vCPU / 4 GiB, 2 vCPU / 8 GiB and 4 vCPU / 16 GiB, with 8 vCPU / 32 GiB or larger by email; the pricing page lists pay-as-you-go machines up to 8 vCPU / 32 GiB. We show 8 vCPU / 32 GiB.","Scale (same rates, higher limits, bring your own cloud) and self-host licences are priced by contact. Self-hosting is free up to 16 vCPUs with no licence key."]},"bestFor":["Dev boxes and agent machines that sit idle much of the day, since idle vCPU is not billed and hibernated machines pay only for disk","Forking a running machine, with memory and processes, to try several approaches in parallel","Teams that want EU hosting or the option to self-host the same platform"],"watchOut":["Machines larger than 4 vCPU / 16 GiB may need a request by email","Only two regions, Frankfurt and New York","A card is required before the first machine, and auto top-up charges it once free credits run out unless turned off","No GPU support is described"],"sources":[{"url":"https://boxd.sh/llms.txt","quote":"vCPU: EUR 0.039 / USD 0.047 per vCPU-hour, on active vCPU only","checked":"2026-11-13"},{"url":"https://boxd.sh/llms.txt","quote":"RAM: EUR 0.012 / USD 0.014 per GiB-hour, on memory actually resident","checked":"2026-11-13"},{"url":"https://boxd.sh/llms.txt","quote":"Disk: EUR 0.00008 / USD 0.000096 per GiB-hour, on disk written to","checked":"2026-11-13"},{"url":"https://boxd.sh/llms.txt","quote":"EUR 30 / USD 30 free credits for new accounts, granted when a payment method is added.","checked":"2026-11-13"},{"url":"https://boxd.sh/llms.txt","quote":"Our own custom VMM, in Rust, built on KVM. No Docker, no external dependencies. Raft consensus for coordination. Hosted in Frankfurt and New York.","checked":"2026-11-13"},{"url":"https://boxd.sh/pricing/","quote":"Usage is metered every second and settled every 15 minutes","checked":"2026-11-13"},{"url":"https://boxd.sh/pricing/","quote":"Up to 8 vCPU · 32 GiB","checked":"2026-11-13"},{"url":"https://boxd.sh/","quote":"Machines come in 1 vCPU / 4 GiB, 2 vCPU / 8 GiB (the default) and 4 vCPU / 16 GiB, and you can pick any of them yourself.","checked":"2026-11-13"},{"url":"https://boxd.sh/","quote":"After 4 hours with no traffic a machine hibernates and pays for disk only, then wakes with its processes intact.","checked":"2026-11-13"},{"url":"https://docs.boxd.sh/","quote":"boxd machines are persistent, hardware-isolated Linux computers that boot in milliseconds, fork, snapshot and hibernate when idle.","checked":"2026-11-13"},{"url":"https://boxd.sh/pricing/","quote":"It is free up to 16 vCPUs, and a licence lifts that limit.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/boxd","score":{"dev-box":{"total":77,"rank":4,"of":38}},"comparablePrice":{"dev-box":16.456},"benchmarks":[]},{"slug":"boxes-dev","name":"boxes.dev","company":"boxes.dev","website":"https://boxes.dev","docsUrl":"https://boxes.dev/help","pricingUrl":"https://boxes.dev/help/plans-and-box-hours","categories":["dev-box"],"tagline":"Cloud Linux devboxes for Claude Code and Codex, cloned from a template snapshot and billed per awake box-hour.","summary":"boxes.dev gives each coding agent its own cloud Linux machine (a devbox) created from a Template box snapshot of your dev environment, including memory. You bring your own Codex or Claude Code subscription or API key; boxes.dev bills only machine time, measured in box-hours, where one box-hour is one hour awake on the standard 4 vCPU / 8 GiB box. Devboxes sleep after about five minutes without activity, keeping files and memory, and sleeping boxes are not billed. Paid seats include a monthly box-hour allowance, with extra usage at $0.40 per box-hour from prepaid credit.","isolation":"Separate Linux machine per devbox, with its own filesystem, processes and ports (virtualization technology not stated)","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["CLI","REST API","Desktop app","iOS and Android apps"],"regions":"Not stated on the pages checked.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.4,"instance":"Standard box: 4 vCPU / 8 GiB RAM / 100 GB disk (one box-hour per awake hour)","storageGibMonth":null,"billingIncrement":"per second","billedWhile":"Only while the box is awake. Sleeping devboxes and Template boxes use no box-hours. A box sleeps after about 5 minutes without activity by default (5, 15, 30 or 60 minutes selectable); a running agent or a keep-awake request keeps it awake.","cpuActiveOnly":false,"unitVcpu":4,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Starter","monthlyFee":19,"includedUsageUsd":20,"limits":"Per paid seat: 50 box-hours a month, 5 awake devboxes plus the Template box, up to 5 projects, email support. Extra usage $0.40 per box-hour from prepaid credit.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":99,"includedUsageUsd":120,"limits":"Per paid seat: 300 box-hours a month, 12 awake devboxes plus the Template box, up to 10 projects, priority support. Extra usage $0.40 per box-hour from prepaid credit.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":249,"includedUsageUsd":300,"limits":"Per paid seat: 750 box-hours a month, 25 awake devboxes plus the Template box, up to 20 projects, priority support. Extra usage $0.40 per box-hour from prepaid credit.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Trial: free, 20 box-hours per user (worth $8 at the extra-usage rate), 3 awake devboxes plus the Template box, up to 2 projects. It is a one-time trial per user that starts at first machine use; some teams must add a payment method first.","notes":["Prices are per box-hour: one hour awake on the standard 4 vCPU / 8 GiB box. The rate shown is the $0.40 extra-usage rate on Starter, Pro and Max, before prepaid discounts (5% at $100, 7.5% at $250, 10% at $500 and above).","Included box-hours are converted to dollars at $0.40 each: Starter 50 = $20, Pro 300 = $120, Max 750 = $300. Allowances are per paid seat and per billing period.","Larger boxes use more box-hours: 4 vCPU / 16 GiB about 1.39, 8 vCPU / 16 GiB 2, 8 vCPU / 24 GiB about 2.39, 8 vCPU / 32 GiB about 2.78 box-hours per awake hour. Memory-heavy sizes therefore cost less than a straight multiple of the 4 vCPU / 8 GiB unit; for example 4 vCPU / 16 GiB is about $0.56 an hour, not $0.80.","The largest size listed is 8 vCPU / 32 GiB; the site has an FAQ about projects that need more memory but the pages checked publish no larger size.","Extra usage is sold as prepaid credit with optional auto-reload, not billed in arrears. A Template box also uses box-hours while awake.","A box awake for 24 hours in a row pauses briefly and usually resumes within seconds if it still has work. Codex or Claude model usage is billed by your own subscription or API account, not by boxes.dev.","Disk is 100 GB on the standard box; no separate storage price is published, and sleeping boxes are described as free.","The Teams offering (SSO, SCIM, audit logs, dedicated resources, volume hourly rates) is priced by contact only."]},"bestFor":["Running several Claude Code or Codex sessions in parallel, each on its own copy of a full dev environment.","Developers who already pay for a Claude or ChatGPT plan and only want to pay for machine time.","Long-lived devboxes that sit idle much of the time, since sleeping boxes keep memory and files and are not billed."],"watchOut":["Compute is sold in fixed box sizes up to 8 vCPU / 32 GiB, not per vCPU or per GiB.","Seat fees are per user, and each seat's included box-hours cannot be shared beyond the team's prepaid credit pool.","The product is built around its agent workbench (desktop, mobile, Slack, Linear); there is no general-purpose sandbox SDK.","Awake time counts even when idle if a keep-awake request or a longer sleep timeout is set."],"sources":[{"url":"https://boxes.dev","quote":"Idle boxes sleep automatically. Sleeping boxes are free. Billed by the second. 4 vCPU / 8 GiB RAM / 100 GB disk standard.","checked":"2026-11-14"},{"url":"https://boxes.dev","quote":"Extra usage starts after your included box-hours, at $0.40 per box-hour on Starter, Pro, and Max before prepaid discounts.","checked":"2026-11-14"},{"url":"https://boxes.dev","quote":"Extra usage is sold as prepaid credit with optional auto-buy · cancel anytime","checked":"2026-11-14"},{"url":"https://boxes.dev","quote":"Take a snapshot—filesystem and memory.","checked":"2026-11-14"},{"url":"https://boxes.dev/help/plans-and-box-hours","quote":"|Starter |$19 per paid seat/month |50 box-hours per user per billing period |5, plus the Template box |Up to 5 |","checked":"2026-11-14"},{"url":"https://boxes.dev/help/plans-and-box-hours","quote":"|Pro |$99 per paid seat/month |300 box-hours per user per billing period |12, plus the Template box |Up to 10 |","checked":"2026-11-14"},{"url":"https://boxes.dev/help/plans-and-box-hours","quote":"|Max |$249 per paid seat/month |750 box-hours per user per billing period |25, plus the Template box |Up to 20 |","checked":"2026-11-14"},{"url":"https://boxes.dev/help/plans-and-box-hours","quote":"|Trial |Free |20 box-hours per user |3, plus the Template box |Up to 2 |","checked":"2026-11-14"},{"url":"https://boxes.dev/help/plans-and-box-hours","quote":"|8 vCPU / 32 GiB |About 2.78 |","checked":"2026-11-14"},{"url":"https://boxes.dev/help/sleep-wake-recover","quote":"A Template box or devbox normally sleeps after about five minutes without activity.","checked":"2026-11-14"},{"url":"https://boxes.dev/help/sleep-wake-recover","quote":"Sleeping keeps the machine's files and memory, like closing a laptop's lid","checked":"2026-11-14"},{"url":"https://boxes.dev/help/data-and-credentials","quote":"Each devbox has its own files, processes, local databases, ports, and threads.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/boxes-dev","score":{"dev-box":{"total":45,"rank":35,"of":38}},"comparablePrice":{"dev-box":87},"benchmarks":[]},{"slug":"browser-use","name":"Browser Use Cloud","company":"Browser Use","website":"https://browser-use.com","docsUrl":"https://docs.browser-use.com","pricingUrl":"https://browser-use.com/pricing","categories":["browser"],"tagline":"Pay-as-you-go cloud browsers and web agents at $0.06 per browser-hour, billed by the minute.","summary":"Browser Use Cloud hosts stealth Chromium browsers that you can drive over CDP with Playwright or Puppeteer, or hand tasks to its hosted web agent. Browsers cost $0.06 per hour, billed by the minute with a one-minute minimum; traffic is billed separately at $5/GB through home-IP proxies or $0.20/GB direct, and CAPTCHA solving is included. There is no subscription: you buy prepaid credits that do not expire, and the number of browsers you can run at once rises with your total spend. New accounts get a one-time $1 credit.","isolation":"Not stated","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","REST API","Playwright","Puppeteer"],"regions":"Not stated for self-serve; EU data residency is offered on Enterprise.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":4,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Per minute, rounded up, with a one-minute minimum","billedWhile":"While the browser session runs; the requested session time is charged upfront and unused time is refunded when the session is stopped","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.06},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"Prepaid credits from $5 (1 credit = $1) that never expire. 10 concurrent browsers to start; 50 after $100 total paid, 250 after $1,000, 500 after $5,000, 1,000 after $25,000. Sessions up to 4 hours.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"One-time $1 signup credit, one per person.","notes":["Traffic is billed on top of browser time: $5/GB through Browser Use's home-IP proxies, $0.20/GB direct or through your own proxy.","CAPTCHA solving is included at no extra charge.","The browser session is charged at the full rate upfront when it starts, and unused time is refunded proportionally when you stop it; billing rounds up to the minute with a one-minute minimum.","Hosted agent tasks are billed separately (for example $0.006 per task for the bu-ultrafast model), plus browser and traffic at the rates above.","Enterprise offers lower committed rates, concurrency past 1,000, EU data residency and on-premises deployment; pricing is custom.","The Browser Use agent library and Browser Harness are MIT-licensed open source; the hosted cloud browsers are not."]},"bestFor":["High-volume browser work at a low hourly rate without a subscription","Bursty workloads, since credits never expire","Teams already using the open-source Browser Use agent library"],"watchOut":["Traffic is metered even without a proxy ($0.20/GB), and home-IP proxy traffic at $5/GB usually costs more than browser time","Only 10 concurrent browsers until you have paid $100 in total","Sessions are limited to 4 hours","The free credit is only $1 and is one-time"],"sources":[{"url":"https://browser-use.com/pricing","quote":"$0.06 per hour, billed by the minute, 1-minute minimum.","checked":"2026-11-13"},{"url":"https://browser-use.com/pricing","quote":"$5/GB on home-IP proxies. $0.20/GB direct or your own proxy.","checked":"2026-11-13"},{"url":"https://browser-use.com/pricing","quote":"Top-ups never expire. The $1 signup credit is one-time.","checked":"2026-11-13"},{"url":"https://browser-use.com/pricing","quote":"By total paid: $100 gets 50, $1,000 gets 250, $5,000 gets 500, $25,000 gets 1,000.","checked":"2026-11-13"},{"url":"https://browser-use.com/pricing","quote":"CAPTCHA solving Included","checked":"2026-11-13"},{"url":"https://docs.browser-use.com/cloud/api-v2/browsers/create-browser-session","quote":"All users: Up to 4 hours per session","checked":"2026-11-13"},{"url":"https://docs.browser-use.com/cloud/api-v2/browsers/create-browser-session","quote":"The full rate is charged upfront when the session starts.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/browser-use","score":{"browser":{"total":42,"rank":9,"of":12}},"comparablePrice":{"browser":0.06},"benchmarks":[]},{"slug":"browserbase","name":"Browserbase","company":"Browserbase","website":"https://www.browserbase.com","docsUrl":"https://docs.browserbase.com","pricingUrl":"https://www.browserbase.com/pricing","categories":["browser"],"tagline":"Hosted headless browsers for agents, sold as monthly plans with included browser hours and per-hour overage.","summary":"Browserbase runs fleets of headless Chromium browsers in the cloud that agents drive with Playwright, Puppeteer, Selenium or its own Stagehand SDK. Each browser runs in a dedicated VM. Plans bundle browser hours, proxy bandwidth, and Search and Fetch API calls; usage past the allowance is billed at $0.12 per browser-hour on Developer and $0.10 on Startup. A free plan includes one browser hour per month with 15-minute sessions.","isolation":"Each browser runs in a dedicated VM, in an isolated subnet","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Node.js","Python","Stagehand","Playwright","Puppeteer","Selenium"],"regions":"us-west-2 (Oregon, default), us-east-1 (Virginia), eu-central-1 (Frankfurt), ap-southeast-1 (Singapore)","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":6,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Per minute, with a one-minute minimum per session","billedWhile":"While a browser session is running","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.1},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1 browser hour per month, 3 concurrent browsers, 15-minute sessions, 5 session creations per minute, no proxies or captcha solving, 3 agent runs, 1,000 Search and 1,000 Fetch calls, $5 of model tokens. No overage.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Developer","monthlyFee":20,"includedUsageUsd":0,"limits":"100 browser hours then $0.12/browser-hour; 1 GB proxy then $12/GB; 25 concurrent browsers; 6-hour sessions; 25 session creations per minute; auto captcha solving; 15 agent runs.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Startup","monthlyFee":99,"includedUsageUsd":0,"limits":"500 browser hours then $0.10/browser-hour; 5 GB proxy then $10/GB; 100 concurrent browsers; 6-hour sessions; 50 session creations per minute; auto captcha solving; 50 agent runs; 10,000 Fetch calls.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: 1 browser hour per month, 3 concurrent browsers and 15-minute sessions, no credit card; usage stops at the allowance.","notes":["The per-hour price shown is the Startup plan overage rate ($0.10/browser-hour, on top of the $99 fee). Developer overage is $0.12/browser-hour. Within the allowance, Developer works out to $0.20/hour and Startup to $0.198/hour if all included hours are used.","Proxies are extra: 1 GB included on Developer then $12/GB, 5 GB on Startup then $10/GB. The free plan has no proxy bandwidth.","Captcha solving is included on paid plans at no listed extra charge; not available on Free.","Fetch API: $1 per 1,000 calls on Developer, $0.50 on Startup, $4 per 1,000 with proxies or Extract, $7 per 1,000 with both. Search overage is listed from $1 to $12 per 1,000 calls depending on the search backend.","Billing is per minute with a one-minute minimum per session, so many very short sessions cost at least a minute each; keep-alive sessions (paid plans) can avoid this.","The Scale plan (250+ concurrent browsers, sessions over 6 hours, HIPAA, SSO) has custom pricing."]},"bestFor":["Teams that want Playwright, Puppeteer or Selenium scripts to run against hosted Chromium with session replays","Workloads that fit inside a monthly bundle of 100 or 500 browser hours","Users of the Stagehand SDK who want browsers, search, fetch and model access on one bill"],"watchOut":["Monthly fee required for anything past 1 browser hour; the free plan stops at its allowance","Proxy traffic is $10-12 per GB beyond the small included amount","Sessions are capped at 6 hours on self-serve plans and 15 minutes on Free","One-minute minimum per session"],"sources":[{"url":"https://www.browserbase.com/pricing","quote":"100 browser hours then $0.12/browser hr","checked":"2026-11-13"},{"url":"https://www.browserbase.com/pricing","quote":"500 browser hours then $0.10/browser hr","checked":"2026-11-13"},{"url":"https://www.browserbase.com/pricing","quote":"Then $12/GB","checked":"2026-11-13"},{"url":"https://docs.browserbase.com/guides/plans-and-pricing","quote":"Browser time is billed by the minute and proxy bandwidth by the MB, rounded, with a one-minute and one-MB minimum per session.","checked":"2026-11-13"},{"url":"https://docs.browserbase.com/guides/plans-and-pricing","quote":"Session Duration | 15 mins | 6 hrs | 6 hrs | 6+ hrs","checked":"2026-11-13"},{"url":"https://docs.browserbase.com/account/enterprise/security","quote":"Each browser runs in a dedicated VM, providing strong isolation at the hardware virtualization layer.","checked":"2026-11-13"},{"url":"https://docs.browserbase.com/optimizations/latency/multi-region","quote":"Available regions: us-west-2 (Oregon), us-east-1 (Virginia), eu-central-1 (Frankfurt), and ap-southeast-1 (Singapore).","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/browserbase","score":{"browser":{"total":54,"rank":5,"of":12}},"comparablePrice":{"browser":0.1},"benchmarks":[]},{"slug":"browserless","name":"Browserless","company":"Browserless","website":"https://www.browserless.io","docsUrl":"https://docs.browserless.io","pricingUrl":"https://www.browserless.io/pricing","categories":["browser"],"tagline":"Hosted Chrome, Firefox and WebKit for Puppeteer, Playwright and BrowserQL, billed in 30-second units on monthly plans.","summary":"Browserless hosts browsers that existing Puppeteer or Playwright scripts connect to over WebSocket, plus BrowserQL, REST APIs for screenshots, PDFs and scraping, and a managed agent API. Plans include a monthly number of units, where one unit is up to 30 seconds of browser time per connection; extra units cost $0.0015 to $0.0020 each, or $0.18 to $0.24 per browser-hour. Residential and datacenter proxies and captcha solves are also paid in units. The source is available under the SSPL or a commercial license and can be self-hosted with Docker.","isolation":"Not stated","isolationKind":"unknown","openSource":false,"selfHost":true,"license":"SSPL-1.0","repoUrl":"https://github.com/browserless/browserless","lastRelease":"2026-11-12","sdks":["TypeScript","Python","BrowserQL","REST API","Puppeteer","Playwright","MCP"],"regions":"San Francisco, London and Amsterdam endpoints; more locations, including the EU, on Enterprise.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":1,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Units of up to 30 seconds per browser connection; each reconnect starts a new unit","billedWhile":"While a browser connection is open","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.18},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1,000 units per month (up to about 8.3 browser-hours); 2 concurrent browsers; 2-minute sessions; residential and datacenter proxies and captcha solving paid from units; no overage.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Prototyping","monthlyFee":35,"includedUsageUsd":0,"limits":"20,000 units per month, then $0.0020 per unit; 5 concurrent browsers (10 if billed annually at $25/month); 15-minute sessions; bring your own proxy; screen recording.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Starter","monthlyFee":200,"includedUsageUsd":0,"limits":"180,000 units per month, then $0.0017 per unit; 30 concurrent browsers (40 if billed annually at $140/month); 30-minute sessions.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":500,"includedUsageUsd":0,"limits":"500,000 units per month, then $0.0015 per unit; 80 concurrent browsers (100 if billed annually at $350/month); 60-minute sessions.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: 1,000 units (30-second blocks) every month, 2 concurrent browsers, 2-minute sessions, no card required.","notes":["A unit is a block of up to 30 seconds of browser time per connection, so one browser-hour is 120 units. The per-hour price shown is the Scale overage rate: $0.0015 x 120 = $0.18. Starter overage is $0.204 per hour and Prototyping $0.24 per hour.","Within the allowance, the monthly-billed plans work out to $0.21 (Prototyping), about $0.133 (Starter) and $0.12 (Scale) per browser-hour if every unit is used; annual billing is 30% cheaper.","Every connection uses at least one unit, and session reconnects count as new connections.","Proxies are paid in units: 6 units per MB for residential and 2 units per MB for datacenter. Each successful captcha solve costs 10 units.","Enterprise (custom pricing) adds private deployments, a commercial self-hosting license, GPU machines and more locations.","The code is dual-licensed under the Server Side Public License or a commercial license; it is source-available rather than OSI open source."]},"bestFor":["Existing Puppeteer or Playwright scripts that only need a new connection URL","Screenshot, PDF and scraping jobs through ready-made REST endpoints","Teams that may later move to a self-hosted deployment"],"watchOut":["Billed per 30-second unit per connection, so many short connections round up","Proxy traffic is paid in units and can use them up quickly","Sessions are capped at 60 minutes even on the largest self-serve plan, and 2 minutes on Free","Commercial self-hosting needs a paid license"],"sources":[{"url":"https://www.browserless.io/pricing","quote":"A “Unit” is a block of browser time of up to 30 seconds per browser connection.","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"$0.0015 per unit overages","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"$0.0020 per unit overages","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"500k units per month","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"Residential proxies (6 units/MB) and datacenter proxies (2 units/MB)","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"10 Units per successful Captcha solve","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"San Francisco, London and Amsterdam endpoints","checked":"2026-11-13"},{"url":"https://www.browserless.io/pricing","quote":"Annual billing saves 30% and includes additional concurrency on the Prototyping, Starter, and Scale plans.","checked":"2026-11-13"},{"url":"https://docs.browserless.io/llms.txt","quote":"write it with BAP, the typed TypeScript and Python SDK","checked":"2026-11-13"},{"url":"https://github.com/browserless/browserless/blob/main/LICENSE","quote":"This work is dual-licensed under the MongoDB Server-Side Public License OR the browserless commercial license.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/browserless","score":{"browser":{"total":35,"rank":12,"of":12}},"comparablePrice":{"browser":0.18},"benchmarks":[]},{"slug":"buildkite","name":"Buildkite hosted agents","company":"Buildkite","website":"https://buildkite.com","docsUrl":"https://buildkite.com/docs/agent/buildkite-hosted","pricingUrl":"https://buildkite.com/pricing","categories":["ci-runner"],"tagline":"Managed Linux, macOS and Windows agents for Buildkite Pipelines, billed per vCPU-minute and metered to the second.","summary":"Buildkite hosted agents run Buildkite Pipelines jobs on Buildkite-managed machines, with each job in an isolated virtualized environment that is destroyed when the job finishes. Linux agents come in Small (2 vCPU, 4 GB), Medium and Large shapes on AMD64 and ARM64, with Mac M4 and Windows agents on Pro. Linux usage costs $0.004 per vCPU-minute, metered to the second with no minimums, and cache volumes add no charge. The Free plan includes 2,000 Linux vCPU-minutes a month; Pro is $30 per active user and includes 4,000.","isolation":"Ephemeral virtualized environment per job with hypervisor-level isolation between clusters; Linux jobs run in a container image inside it","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Pipeline YAML","Buildkite agent","REST and GraphQL APIs"],"regions":"Hosted agents run only in US East Coast data centers.","features":{"pauseResume":false,"snapshots":null,"gpu":null,"customImages":true,"networkControls":null,"maxSessionHours":8,"maxVcpu":8,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.24,"gibHour":null,"instanceHour":0.48,"instance":"Linux Small: 2 vCPU, 4 GB, AMD64 or ARM64 ($0.008 per minute)","storageGibMonth":null,"billingIncrement":"per second, in vCPU-minutes, no minimum","billedWhile":"Only while a job runs; startup and shutdown are excluded. Cache volumes are not charged.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Up to 2,000 Linux AMD64 vCPU-minutes a month on the Small shape only, up to 20 Linux vCPUs at once, up to 10 concurrent jobs, 5 users. No ARM64, Mac or Windows hosted agents.","includedVcpuHours":33.333333,"includedGibHours":0,"capped":true},{"name":"Pro","monthlyFee":30,"includedUsageUsd":0,"limits":"$30 per active user a month, up to 50 users. Includes 4,000 Linux AMD64 vCPU-minutes a month, then per-minute rates. Up to 256 Linux AMD64 vCPUs, 64 ARM64, 48 Mac and 64 Windows vCPUs at once; Small, Medium and Large shapes. 10 self-hosted agents included, then $3.50 per agent a month.","includedVcpuHours":66.666667,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan with up to 2,000 Linux vCPU-minutes a month (about 16.7 hours on the 2 vCPU Small shape). New organizations also get a 30-day all-access trial with no credit card.","notes":["Linux rate is $0.004 per vCPU-minute x 60 = $0.24 per vCPU-hour. Small (2 vCPU, 4 GB) is $0.008 per minute ($0.48 per hour), Medium (4 vCPU, 16 GB) $0.016 and Large (8 vCPU, 32 GB) $0.032. Price depends on vCPUs only; memory is not billed separately.","AMD64 and ARM64 shapes cost the same. Mac M4 agents are $0.02 per vCPU-minute (6 vCPU Medium $0.12 per minute) and Windows $0.008 per vCPU-minute (2 vCPU Small $0.016 per minute); both are pay-as-you-go on Pro.","The Pro fee is per active user: anyone who logs in, uses the API or triggers a build in a month.","Included allowances are counted in vCPU-minutes: 2,000 on Free (33.3 vCPU-hours) and 4,000 on Pro (66.7 vCPU-hours), for Linux AMD64 only.","Jobs can run up to 8 hours; longer runs need support. Extra Large and bigger shapes (up to 64 vCPU / 256 GB) and remote Docker builders are Enterprise only.","Cache volumes, git mirrors and an internal container registry are included at no extra charge. Artifact storage beyond the plan allowance is $0.05 per GB-month on Pro.","Memory is stated in GB; we treat it as GiB."]},"bestFor":["Teams already on Buildkite Pipelines that want to stop running their own agents","Mixed fleets that combine hosted agents with self-hosted agents in one pipeline","Short jobs, since billing is metered to the second with no minimum"],"watchOut":["Pro is charged per active user on top of compute","The Free plan offers only the 2 vCPU / 4 GB Linux shape","Hosted agents run only in US East Coast data centers","Only works with Buildkite Pipelines, not GitHub Actions"],"sources":[{"url":"https://buildkite.com/pricing","quote":"$0.004 USD per vCPU minute","checked":"2026-11-13"},{"url":"https://buildkite.com/pricing","quote":"Usage is billed per vCPU minute and metered to the second, with no minimums and no rounding.","checked":"2026-11-13"},{"url":"https://buildkite.com/pricing","quote":"Up to 2,000 Linux vCPU minutes per month","checked":"2026-11-13"},{"url":"https://buildkite.com/pricing","quote":"Includes 4,000 Linux AMD64 vCPU minutes per month","checked":"2026-11-13"},{"url":"https://buildkite.com/pricing","quote":"$ 30 per active user/month","checked":"2026-11-13"},{"url":"https://buildkite.com/pricing","quote":"Every shape is available on AMD64 and ARM64 at the same rate.","checked":"2026-11-13"},{"url":"https://buildkite.com/docs/agent/buildkite-hosted","quote":"Startup and shutdown periods are excluded, with no minimum charges.","checked":"2026-11-13"},{"url":"https://buildkite.com/docs/agent/buildkite-hosted","quote":"Is exclusively hosted in US East Coast data centers","checked":"2026-11-13"},{"url":"https://buildkite.com/docs/agent/buildkite-hosted","quote":"There are no supplementary charges for storage or cache usage.","checked":"2026-11-13"},{"url":"https://buildkite.com/docs/agent/buildkite-hosted/linux","quote":"Buildkite employs a multi-tenant architecture, where each job runs in a completely isolated virtualized environment.","checked":"2026-11-13"},{"url":"https://buildkite.com/docs/agent/buildkite-hosted/linux","quote":"To accommodate different workloads, instances are capable of running up to 8 hours.","checked":"2026-11-13"},{"url":"https://buildkite.com/docs/agent/buildkite-hosted/linux","quote":"The Free plan only provides access to the small-sized AMD64 instance shape.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/buildkite","score":{"ci-runner":{"total":53,"rank":10,"of":10}},"comparablePrice":{"ci-runner":78},"benchmarks":[]},{"slug":"claude-code-web","name":"Claude Code on the web","company":"Anthropic","website":"https://claude.ai/code","docsUrl":"https://code.claude.com/docs/en/claude-code-on-the-web","pricingUrl":"https://claude.com/pricing","categories":["coding-agent"],"tagline":"Claude Code sessions in Anthropic-managed cloud VMs, included in paid Claude plans with no separate VM charge.","summary":"Claude Code cloud sessions run in an isolated Ubuntu VM that Anthropic manages, started from claude.ai/code, the Claude apps, the desktop app or the terminal with claude --cloud. Each VM has about 4 vCPUs, 16 GB of RAM and 30 GB of disk, pauses when idle with its files saved, and starts from a cached snapshot of your setup script. There is no separate charge for the VM: sessions count toward the plan's shared Claude usage limits, available from the $20 Pro plan upward. Network access is limited by default and can be widened or turned off per environment.","isolation":"Isolated Anthropic-managed VM per session","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Web app","Desktop and mobile apps","CLI"],"regions":"Not stated.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":false,"networkControls":true,"maxSessionHours":null,"maxVcpu":4,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"No separate machine billing; sessions use the plan's usage limits","billedWhile":"The VM is not billed. Model usage in a cloud session counts toward the same five-hour and weekly limits as all other Claude and Claude Code use.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pro","monthlyFee":20,"includedUsageUsd":0,"limits":"Claude Code including cloud sessions. Usage limits every 5 hours and every week, shared across Claude chat, Claude Code and cloud sessions. $17/month when billed annually ($200 up front). Usage credits at API rates after the limit.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":100,"includedUsageUsd":0,"limits":"5x Pro usage every 5 hours and every week, priority access at high traffic times, plus $100 a month in Claude Platform API credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Ultra","monthlyFee":200,"includedUsageUsd":0,"limits":"10x Pro weekly usage with no 5-hour limit for verified accounts, plus $200 a month in Claude Platform API credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"None. The Free Claude plan does not include Claude Code; cloud sessions need Pro, Max, Ultra, Team or an eligible Enterprise seat.","notes":["The cloud machine is not priced separately: Anthropic states there is no separate charge for the cloud VM. Cost is the plan fee plus any usage credits bought at API rates after the plan limits.","Running several cloud sessions in parallel uses the shared rate limits proportionately faster.","Resource ceilings per session (approximate, may change): 4 vCPUs, 16 GB of RAM, 30 GB of disk. We read 16 GB as 16 GiB. Tasks needing much more memory may be stopped.","No fixed session length is stated. After a few minutes without activity the VM pauses with files saved; a paused VM can later be reclaimed, and reopening the session provisions a fresh VM with the conversation restored but not running processes.","A setup script that finishes in roughly five minutes is cached as a filesystem snapshot that later sessions start from; the cache expires after about seven days. Replacing the base image is not supported; Docker is available inside the VM.","Network access is limited to an allowlist by default and can be set per environment; GitHub credentials stay outside the VM behind a proxy.","Team and Enterprise plans also include cloud sessions (Enterprise needs premium or Chat + Claude Code seats). Enterprise is listed at $20 per seat per month plus usage at API rates; Team pricing is not on the individual pricing page checked. Organizations can route sessions to their own self-hosted environments."]},"bestFor":["Claude subscribers who want to hand off coding tasks to the cloud at no extra machine cost.","Moving work between terminal and cloud with claude --cloud and --teleport.","Watching pull requests and auto-fixing CI failures and review comments."],"watchOut":["Each session is limited to about 4 vCPUs and 16 GB of RAM.","Cloud sessions share rate limits with all other Claude usage on the account.","Repository cloning and pull requests in hosted environments require GitHub.","Paused VMs can be reclaimed, losing running processes."],"sources":[{"url":"https://code.claude.com/docs/en/claude-code-on-the-web","quote":"It counts toward your plan's usage limits alongside the rest of your Claude and Claude Code usage, and there's no separate charge for the cloud VM.","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/claude-code-on-the-web","quote":"Cloud sessions are available on Pro, Max, Ultra, and Team plans, and for Enterprise users with premium seats or Chat + Claude Code seats.","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/claude-code-on-the-web","quote":"each session runs in an isolated, Anthropic-managed VM","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/claude-code-on-the-web","quote":"Running multiple tasks in parallel consumes more rate limits proportionately.","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/cloud-environments","quote":"Cloud sessions in Anthropic-hosted environments run with approximate resource ceilings that may change over time:","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/cloud-environments","quote":"16 GB of RAM","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/cloud-environments","quote":"after a few minutes without activity, a session's VM pauses with its files saved, and a paused VM can later be reclaimed","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/cloud-environments","quote":"Replacing the base image entirely isn't supported yet.","checked":"2026-11-14"},{"url":"https://code.claude.com/docs/en/cloud-environments","quote":"When setup completes within roughly five minutes, Anthropic snapshots the filesystem and reuses that snapshot as the starting point for later sessions.","checked":"2026-11-14"},{"url":"https://claude.com/pricing","quote":"Per month with annual subscription discount ( $200 billed up front). $20 if billed monthly.","checked":"2026-11-14"},{"url":"https://claude.com/pricing","quote":"5x Pro usage every 5 hours and every week","checked":"2026-11-14"},{"url":"https://claude.com/pricing","quote":"Claude Code is included in all paid plans.","checked":"2026-11-14"},{"url":"https://claude.com/pricing","quote":"$200 a month in API credits, up from $100 on Max","checked":"2026-11-14"},{"url":"https://claude.com/pricing","quote":"Pricing is $20 per seat per month plus usage billed at API rates","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/claude-code-web","score":{"coding-agent":{"total":73,"rank":3,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"cloudflare-dynamic-workers","name":"Cloudflare Dynamic Workers","company":"Cloudflare","website":"https://developers.cloudflare.com/dynamic-workers/","docsUrl":"https://developers.cloudflare.com/dynamic-workers/","pricingUrl":"https://developers.cloudflare.com/dynamic-workers/pricing/","categories":["code-sandbox"],"tagline":"V8 isolates created at runtime from a Worker to run untrusted JavaScript or WebAssembly, billed on CPU time.","summary":"Dynamic Workers let a Cloudflare Worker load and run code supplied at runtime in a new V8 isolate, as a lightweight alternative to containers. The parent Worker decides which bindings the code receives and can block, filter or rewrite its outbound requests. Billing uses Workers Standard rates for requests and CPU milliseconds (including isolate startup), plus a charge per unique Dynamic Worker created each day. It runs JavaScript and WebAssembly only, with 128 MB of memory per isolate.","isolation":"V8 isolate per Dynamic Worker","isolationKind":"isolate","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript (Worker Loader binding inside a Worker)","WebAssembly"],"regions":"Runs on Cloudflare's global network; region choice is not described.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":false,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":0.125},"pricing":{"rates":{"vcpuHour":0.072,"gibHour":0,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per CPU millisecond, plus per request and per unique Dynamic Worker per day","billedWhile":"CPU time only: isolate startup plus active execution, excluding time waiting on I/O. Wall-clock duration and memory are not billed.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Workers Paid","monthlyFee":5,"includedUsageUsd":0,"limits":"Required for Dynamic Workers. Includes per month 30 million CPU milliseconds, 10 million requests and 1,000 unique Dynamic Workers; then $0.02 per million CPU ms, $0.30 per million requests and $0.002 per Dynamic Worker per day.","includedVcpuHours":8.3333,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"None. Dynamic Workers are only available on the Workers Paid plan ($5/month minimum), which includes a monthly usage allowance.","notes":["CPU time is $0.02 per million CPU milliseconds; one CPU-hour is 3.6 million ms, so $0.072 per CPU-hour of actual execution. Memory is not billed separately, so we show $0 for memory.","Unlike standard Workers, Dynamic Workers also bill the CPU time spent starting the isolate and parsing the code.","Each unique Dynamic Worker (same ID and code) created in a day is counted; beyond 1,000 per month it costs $0.002 per Dynamic Worker per day. Loading without a stable ID counts every invocation as a new one.","Requests cost $0.30 per million beyond 10 million per month; RPC calls into a Dynamic Worker count as requests.","The plan allowance of 30 million CPU ms equals about 8.33 CPU-hours.","Limits per isolate: 128 MB memory (shown as 0.125 GiB) and up to 5 minutes of CPU time per invocation (default 30 seconds). No duration limit is stated for wall-clock time.","Only JavaScript and WebAssembly run; there is no filesystem image or general Linux environment."]},"bestFor":["Running LLM-written JavaScript against an API you expose, instead of many tool calls","Very high volumes of short executions where millisecond startup matters","Platforms on Cloudflare that run code uploaded by their users"],"watchOut":["JavaScript and WebAssembly only; no Python, shell or Linux processes","128 MB memory per isolate and up to 5 minutes of CPU per invocation","Requires the Workers Paid plan; there is no free use","Loading code without a stable ID bills a new Dynamic Worker on every call"],"sources":[{"url":"https://developers.cloudflare.com/dynamic-workers/pricing/","quote":"| **CPU time** ¹ | 30 million CPU milliseconds per month | +$0.02 per million CPU milliseconds |","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/dynamic-workers/pricing/","quote":"| **Dynamic Workers created daily** | 1,000 unique Dynamic Workers per month | +$0.002 per Dynamic Worker per day |","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/dynamic-workers/pricing/","quote":"| **Requests** ¹ | 10 million per month | +$0.30 per million requests |","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/dynamic-workers/pricing/","quote":"Dynamic Workers are currently only available on the Workers Paid plan.","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/dynamic-workers/pricing/","quote":"**Startup time**: The compute required to initialize the isolate and parse your code.","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/workers/platform/pricing/","quote":"for a minimum charge of $5 USD per month for an account","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/workers/platform/limits/","quote":"| Memory per isolate | 128 MB |","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/workers/platform/limits/","quote":"| CPU time per HTTP request | 10 ms | 5 min (default: 30 seconds) |","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/dynamic-workers/","quote":"Use Dynamic Workers as a lightweight alternative to containers for securely sandboxing code you do not trust.","checked":"2026-11-13"},{"url":"https://developers.cloudflare.com/dynamic-workers/usage/egress-control/","quote":"Set `globalOutbound` to `null` to fully isolate the dynamic Worker from the network","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/cloudflare-dynamic-workers","score":{"code-sandbox":{"total":48,"rank":59,"of":63}},"comparablePrice":{"code-sandbox":12.0032024},"benchmarks":[]},{"slug":"cloudflare-sandbox","name":"Cloudflare Sandbox SDK","company":"Cloudflare","website":"https://developers.cloudflare.com/sandbox/","docsUrl":"https://developers.cloudflare.com/sandbox/","pricingUrl":"https://developers.cloudflare.com/containers/platform/pricing/","categories":["code-sandbox"],"tagline":"Linux sandboxes on Cloudflare Containers, started and controlled from a Worker and a Durable Object.","summary":"Cloudflare's Linux sandboxes are Containers instances that a Worker starts through a Durable Object; each instance is a microVM with its own kernel running an image you provide. Billing follows Containers pricing: active CPU time plus provisioned memory and disk, metered every 10 ms while the instance runs, on top of the $5/month Workers Paid plan. Instances stop after an inactivity timeout of up to 6 hours, and files can be carried to the next instance with container snapshots (public beta).","isolation":"microVM per container instance (own kernel and network)","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript (Workers / @cloudflare/sandbox)"],"regions":"Runs on Cloudflare's network; the pages checked do not document choosing a region.","features":{"pauseResume":false,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":4,"maxMemoryGib":12},"pricing":{"rates":{"vcpuHour":0.072,"gibHour":0.009,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per 10 ms","billedWhile":"While the container instance runs: CPU only for active usage; memory and disk for the provisioned instance size, even when idle.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Workers Paid","monthlyFee":5,"includedUsageUsd":0,"limits":"Includes 375 vCPU-minutes, 25 GiB-hours of memory and 200 GB-hours of disk per month (allowances, not a dollar credit). Default per-instance max 4 vCPU, 12 GiB memory, 20 GB disk; account max 15,000 concurrent vCPU and 60 TiB concurrent memory.","includedVcpuHours":6.25,"includedGibHours":25,"capped":false}],"freeTierKind":"none","freeTier":"None for sandboxes: Containers require the Workers Paid plan ($5/month), which includes 375 vCPU-minutes, 25 GiB-hours of memory and 200 GB-hours of disk each month.","notes":["Rates converted from per-second list prices: $0.000020/vCPU-s = $0.072/vCPU-hour; $0.0000025/GiB-s = $0.009/GiB-hour; disk $0.00000007/GB-s = $0.000252/GB-hour.","CPU is billed on active usage only; memory and disk are billed on the provisioned instance type for as long as the instance runs, so an idle running instance still costs memory and disk.","Disk is not a persistent volume: it is billed per GB-second only while the instance runs, so we show no monthly storage rate. Snapshots are free during the public beta (pricing to be announced at general availability) and expire 30 days after creation or last restore.","Workers Paid included usage is worth about $0.73/month at list rates (375 vCPU-min = $0.45, 25 GiB-h = $0.225, 200 GB-h disk = $0.0504); it is a resource allowance, not a dollar credit, and our calculator subtracts it from usage.","Egress: $0.025/GB in North America & Europe (1 TB/month included), $0.05/GB Oceania, Korea, Taiwan (500 GB included), $0.04/GB elsewhere (500 GB included).","Each sandbox also uses a Worker and a Durable Object, which are billed at Workers and Durable Objects rates on top of Containers usage.","Instance types are fixed sizes (lite 1/16 vCPU 256 MiB up to standard-4 4 vCPU 12 GiB 20 GB) or custom sizes; per-instance maximums are defaults that can be raised on request.","Session length: after the Durable Object becomes inactive, the instance keeps running for at most the inactivity timeout (up to 6 hours); Cloudflare does not guarantee an instance runs for a set time, so we show no maximum session."]},"bestFor":["Apps already on Cloudflare Workers that need a full Linux environment per user or task","Bursty agent workloads that can stop instances quickly, since CPU is billed only when active","Running untrusted code with internet access turned off or outbound traffic routed through the Worker"],"watchOut":["Requires the $5/month Workers Paid plan; no free tier for Containers","Idle running instances still pay for provisioned memory and disk","Default instance ceiling is 4 vCPU and 12 GiB memory","No in-memory pause: instances stop after an inactivity timeout of up to 6 hours, and snapshots (public beta) restore files but not processes"],"sources":[{"url":"https://developers.cloudflare.com/containers/platform/pricing/","quote":"Containers are billed for every 10ms that they are actively running ... 375 vCPU-minutes/month + $0.000020 per additional vCPU-second ... 25 GiB-hours/month included +$0.0000025 per additional GiB-second","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/containers/platform/pricing/","quote":"Memory and disk usage are based on the provisioned resources for the instance type you select, while CPU usage is based on active usage only.","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/containers/platform/pricing/","quote":"Snapshots are available at no charge during the public beta.","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/workers/platform/pricing/","quote":"The Workers Paid plan includes Workers, Pages Functions, Workers KV, Hyperdrive, and Durable Objects usage for a minimum charge of $5 USD per month for an account.","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/containers/platform/limits/","quote":"vCPU per instance | 4 ... Memory per instance | 12 GiB","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/sandbox/","quote":"Each instance is a microVM with its own kernel and network, so no other workload shares it.","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/sandbox/concepts/lifetime/","quote":"After the Durable Object becomes inactive, the instance keeps running for the time set with setInactivityTimeout(), up to 6 hours.","checked":"2026-11-14"},{"url":"https://developers.cloudflare.com/containers/platform/limits/","quote":"Snapshot retention | 30 days from creation or the most recent restore","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/cloudflare-sandbox","score":{"code-sandbox":{"total":56,"rank":48,"of":63}},"comparablePrice":{"code-sandbox":18.2642},"benchmarks":[{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":107110,"unit":"ms","better":"lower","date":"2026-11-13","rank":24,"of":25}]},{"slug":"coder","name":"Coder","company":"Coder Technologies","website":"https://coder.com","docsUrl":"https://coder.com/docs","pricingUrl":"https://coder.com/pricing","categories":["dev-box"],"tagline":"Self-hosted platform for Terraform-defined development workspaces and AI coding agents on your own infrastructure.","summary":"Coder is an AGPL-3.0 platform you deploy yourself to provision development workspaces for people and coding agents. Workspaces are defined in Terraform templates, so they can be VMs, Kubernetes pods or Docker containers, and idle ones shut down automatically. The free Community edition has unlimited workspaces and templates; Premium and AI Premium add enterprise access control, high availability, support and unlimited Coder Agents, priced per user on request.","isolation":"Depends on the template: cloud VMs, Kubernetes pods or Docker containers provisioned with Terraform","isolationKind":"unknown","openSource":true,"selfHost":true,"license":"AGPL-3.0","repoUrl":"https://github.com/coder/coder","lastRelease":"2026-11-10","sdks":["CLI","REST API","Terraform provider","VS Code and JetBrains plugins"],"regions":"Wherever you deploy it: your own cloud accounts, Kubernetes clusters or servers.","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software; you pay your own infrastructure costs","billedWhile":"Not applicable: workspaces cost whatever your cloud or servers charge","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Community edition is free and open source (AGPL-3.0) with unlimited workspaces and templates, up to 5 active Coder Agents and 1 platform integration.","notes":["Premium is billed annually per user and AI Premium is custom; neither has a published price.","Workspace stop and start come from autostart/autostop schedules; what is preserved depends on the template.","Isolation depends entirely on the infrastructure your templates create."]},"bestFor":["Companies that must keep developer and agent environments inside their own cloud or data center","Standardizing workspaces as Terraform templates across VMs, Kubernetes and Docker","Running Claude Code, Codex and other agents in governed, auto-stopping workspaces"],"watchOut":["Paid tiers have no public price","You operate the control plane, database and infrastructure","Isolation strength depends on the template you choose (container or VM)"],"sources":[{"url":"https://github.com/coder/coder","quote":"Coder is a self-hosted platform for cloud development environments and AI coding agents.","checked":"2026-11-13"},{"url":"https://github.com/coder/coder","quote":"EC2 VMs, Kubernetes Pods, Docker Containers, etc.","checked":"2026-11-13"},{"url":"https://github.com/coder/coder","quote":"Automatically shutdown idle resources to save on costs","checked":"2026-11-13"},{"url":"https://coder.com/pricing","quote":"For hobbyists and small teams ready to join Coder's open-source community.","checked":"2026-11-13"},{"url":"https://coder.com/pricing","quote":"Active Coder Agents | Up to 5 | Up to 5 | Unlimited","checked":"2026-11-13"},{"url":"https://coder.com/pricing","quote":"Platform integrations (e.g., GitHub, Jfrog, Backstage) | Limited to 1 integration | Unlimited | Unlimited","checked":"2026-11-13"},{"url":"https://github.com/coder/coder/releases/tag/v2.38.7","quote":"v2.38.7","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/coder","score":{"dev-box":{"total":50,"rank":31,"of":38}},"comparablePrice":{"dev-box":null},"benchmarks":[]},{"slug":"codesandbox","name":"CodeSandbox","company":"CodeSandbox (part of Together AI)","website":"https://codesandbox.io","docsUrl":"https://codesandbox.io/docs/sdk","pricingUrl":"https://codesandbox.io/pricing","categories":["code-sandbox","dev-box"],"tagline":"MicroVM sandboxes with memory snapshots, used through the CodeSandbox SDK or its cloud editor, billed in VM credits.","summary":"CodeSandbox runs VM Sandboxes as microVMs that use memory snapshotting to start, resume and fork in seconds, either from the CodeSandbox SDK or as cloud development environments in its browser editor and VS Code. Runtime is billed in VM credits by machine size, from 5 credits ($0.075) an hour for Pico to 320 credits ($4.80) an hour for XLarge, metered per minute. The free Build plan includes 400 credits a month and 10 concurrent VMs; the Scale plan from $170 a month raises concurrency to 250 VMs and sizes to 16 vCPU. CodeSandbox is part of Together AI.","isolation":"MicroVM per VM Sandbox, with memory snapshots for resume and fork","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript"],"regions":"Not stated on the pages checked.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.075,"instance":"Pico: 1 core / 2 GB (5 credits per hour)","storageGibMonth":null,"billingIncrement":"per minute (rounded up)","billedWhile":"VM runtime while a VM Sandbox is running; hibernated sandboxes are resumed from memory snapshots.","cpuActiveOnly":false,"unitVcpu":1,"unitGib":2,"browserHour":null},"gpu":[],"plans":[{"name":"Build","monthlyFee":0,"includedUsageUsd":6,"limits":"400 VM credits a month (about 40 hours of Nano), 5 members, 10 concurrent VM Sandboxes, 20 new SDK sandboxes an hour, 1,000 SDK requests an hour, VMs up to 4 vCPU / 8 GiB, 20 GB storage per VM.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":170,"includedUsageUsd":0,"limits":"From $170 a month per workspace. Up to 20 members, 250 concurrent VMs, 1,000 new sandboxes an hour, 10,000 requests an hour, VMs up to 16 vCPU / 32 GiB, on-demand credits. The included VM credits are stated as 160 hours a month on the pricing page and 1,100 credits in the SDK docs.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Build plan: free, with 400 VM credits each month (unused credits do not roll over) and up to 10 concurrent VMs.","notes":["VM credits cost $0.015 each on the pricing page ($0.01486 in the docs). Per-size prices on the pricing page are linear at $0.075 per hour for each 1 core / 2 GB: Pico 1/2 GB $0.075, Nano 2/4 GB $0.15, Micro 4/8 GB $0.30, Small 8/16 GB $0.60, Medium 16/32 GB $1.20, Large 32/64 GB $2.40, XLarge 64/128 GB $4.80.","The docs list slightly lower prices ($0.0743 per hour for Pico, $0.1486 for Nano) and describe Pico as 2 cores / 1 GB; we use the pricing page.","CodeSandbox states RAM in GB and GiB in different places; we treat them as GiB.","Billing uses minutes as the smallest unit: 3 minutes 25 seconds bills as 4 minutes.","Build plan's 400 credits are worth $6 at $0.015 per credit (our conversion). Scale's included credits are stated inconsistently (160 hours vs 1,100 credits), so we show no included dollar amount.","Session length is unlimited on every plan. Enterprise offers VMs up to 64 vCPU / 128 GiB, bulk credit discounts and dedicated clusters.","Storage pricing beyond the 20 GB per VM is not stated. Network controls and GPU support are not described on the pages checked."]},"bestFor":["Products that embed a browser IDE or per-user dev environments","Workloads that resume or fork running VMs from memory snapshots","Trying an SDK sandbox for free with a monthly credit allowance"],"watchOut":["The free plan caps VMs at 4 vCPU / 8 GiB and 10 concurrent","Billing rounds up to whole minutes","Prices and included credits differ between the pricing page and the docs","Only a JavaScript/TypeScript SDK is documented"],"sources":[{"url":"https://codesandbox.io/pricing","quote":"VM credits are only used on VM Sandboxes to translate virtual machine (VM) runtime per hour into precise costs. They are priced at $0.015 each.","checked":"2026-11-14"},{"url":"https://codesandbox.io/pricing","quote":"priced at $0.15 per hour","checked":"2026-11-14"},{"url":"https://codesandbox.io/pricing","quote":"160 hours of monthly VM credits","checked":"2026-11-14"},{"url":"https://codesandbox.io/pricing","quote":"Programmatically create and manage Sandboxes at scale, but limited to 10 concurrent VM Sandboxes.","checked":"2026-11-14"},{"url":"https://codesandbox.io/docs/sdk/pricing","quote":"We use minutes as the smallest unit of measurement for VM credits.","checked":"2026-11-14"},{"url":"https://codesandbox.io/docs/sdk/pricing","quote":"Your Scale plan includes 1100 free VM credits each month","checked":"2026-11-14"},{"url":"https://codesandbox.io/docs/learn/credit-usage/credits","quote":"A workspace on the free plan gets 400 credits per month.","checked":"2026-11-14"},{"url":"https://codesandbox.io/docs/learn/vm-sandboxes/overview","quote":"You can fork VM Sandbox within 2 seconds and continue with an exact clone of the microVM.","checked":"2026-11-14"},{"url":"https://codesandbox.io/docs/learn/vm-sandboxes/overview","quote":"VM Sandboxes run VMs and leverage our memory snapshotting to spin up an environment in 1.5 seconds.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/codesandbox","score":{"code-sandbox":{"total":69,"rank":17,"of":63},"dev-box":{"total":68,"rank":13,"of":38}},"comparablePrice":{"code-sandbox":26.4,"dev-box":33},"benchmarks":[]},{"slug":"containarium","name":"Containarium","company":"FootprintAI","website":"https://containarium.dev","docsUrl":"https://github.com/FootprintAI/Containarium","pricingUrl":"https://containarium.dev/pricing","categories":["dev-box"],"tagline":"Open-source runtime giving each agent a persistent SSH box on LXC or Kubernetes; hosted seats $20/month.","summary":"Containarium is an Apache-2.0 agent runtime that gives each agent a persistent, SSH-reachable Linux box in an unprivileged container, managed through a CLI and MCP servers. It runs on LXC or Kubernetes, offers an opt-in per-tenant eBPF egress policy, a model gateway that keeps provider keys out of the box, and GPU passthrough. Self-hosting is free on your own Ubuntu VM; a hosted cloud sells a $20 per month seat with a fixed CPU, memory, disk and transfer allowance and no overage billing.","isolation":"Unprivileged LXC container (or Kubernetes pod) per box; tenant-to-tenant isolation is not on by default","isolationKind":"container","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/FootprintAI/Containarium","lastRelease":"2026-11-12","sdks":["CLI","MCP server","SSH"],"regions":"Self-hosted: your own Ubuntu 24.04 VM or Kubernetes cluster. Hosted cloud region not stated.","features":{"pauseResume":null,"snapshots":null,"gpu":true,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Self-hosted: not applicable. Hosted: flat monthly seat, not metered","billedWhile":"Self-hosted: no charges. Hosted: flat per-seat monthly price regardless of usage","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Seat (hosted)","monthlyFee":20,"includedUsageUsd":0,"limits":"One box per seat: 2 vCPU (1 guaranteed, bursts to 2 when the host has capacity), 4 GiB RAM, 20 GB disk, 100 GB transfer per month. No metering and no overage charges; idle boxes sleep. Marked as launch pricing.","includedVcpuHours":0,"includedGibHours":0,"capped":true}],"freeTierKind":"recurring","freeTier":"Self-hosting is free under Apache-2.0. The hosted cloud has no free tier listed.","notes":["The hosted seat is $20 per month flat, $0.0273972602739726 per hour when divided over a 730-hour month, for a shared 2 vCPU / 4 GiB box.","The pricing page calls this launch pricing and says the number may change.","Teams, dedicated hardware and runs in your own cloud are priced by contacting sales.","The project's own security notes say tenant-to-tenant isolation is not on by default; eBPF egress policy is opt-in."]},"bestFor":["Giving coding agents a persistent Linux box they reach over SSH and MCP","Self-hosting many agent boxes on one Ubuntu VM or a Kubernetes cluster","Users who want a flat monthly price with no usage bill on the hosted cloud"],"watchOut":["Container isolation, and tenant-to-tenant isolation is off by default","Hosted seat is a shared box with only 1 guaranteed vCPU","Hosted price is launch pricing and may change"],"sources":[{"url":"https://github.com/FootprintAI/Containarium","quote":"Each agent gets a persistent, SSH-reachable box in an unprivileged container, with an opt-in per-tenant eBPF egress policy","checked":"2026-11-13"},{"url":"https://github.com/FootprintAI/Containarium","quote":"Tenant-to-tenant isolation is not on by default","checked":"2026-11-13"},{"url":"https://github.com/FootprintAI/Containarium","quote":"Open-source agent runtime — SSH-native isolation, eBPF egress policy, Kubernetes + LXC backends, GPU passthrough, MCP-native CLI","checked":"2026-11-13"},{"url":"https://containarium.dev/pricing","quote":"$20 / seat / month","checked":"2026-11-13"},{"url":"https://containarium.dev/pricing","quote":"2 vCPU (1 guaranteed, bursts to 2 when the host has capacity) · 4 GiB RAM","checked":"2026-11-13"},{"url":"https://containarium.dev/pricing","quote":"We do not meter by the second and we do not charge for going over.","checked":"2026-11-13"},{"url":"https://containarium.dev/pricing","quote":"The product is young and the number may move as we learn what people run.","checked":"2026-11-13"},{"url":"https://github.com/FootprintAI/Containarium/releases/tag/v0.136.0","quote":"v0.136.0","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/containarium","score":{"dev-box":{"total":48,"rank":32,"of":38}},"comparablePrice":{"dev-box":null},"benchmarks":[]},{"slug":"createos","name":"CreateOS Sandbox","company":"NodeOps, Inc.","website":"https://createos.sh/products/sandbox","docsUrl":"https://createos.sh/docs/Sandbox/Overview/","pricingUrl":"https://createos.sh/pricing/sandbox","categories":["code-sandbox"],"tagline":"Firecracker microVMs billed per second, with no egress fees and pause that stops vCPU billing.","summary":"CreateOS Sandbox runs each workload in a Firecracker microVM with its own kernel, created from a TypeScript SDK, CLI, MCP server or REST API in EU or US regions. Every plan pays the same per-second rate: $0.03616363 per vCPU-hour, $0.01159025 per GiB-hour of memory and $0.0001584 per GB-hour of disk, with no egress charges. Pausing snapshots memory and device state and stops vCPU billing, but memory and storage keep billing while paused. Plans only set limits such as concurrent sandboxes and the largest size.","isolation":"Firecracker microVM per sandbox with its own guest kernel; egress rules enforced on the host outside the VM","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","CLI","MCP server","REST API"],"regions":"EU and US regions, chosen per sandbox. Running on your own infrastructure is available to enterprise customers only.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":8},"pricing":{"rates":{"vcpuHour":0.03616363,"gibHour":0.01159025,"instanceHour":null,"instance":null,"storageGibMonth":0.115632,"billingIncrement":"per second","billedWhile":"vCPU, memory and disk while running. Paused sandboxes are still billed for memory and disk, but not vCPU. Disk bills for as long as the sandbox exists.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1 concurrent sandbox, 10 per day, largest shape 1 vCPU / 1 GB, 10 GiB disk, no extra disks or templates. 500 credits at signup.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Beginner","monthlyFee":10,"includedUsageUsd":0,"limits":"From $10 to $50 a month, with 1,000 to 5,000 initial credits and a 1.2x top-up bonus. 5 concurrent sandboxes, 50 per day, largest shape 4 vCPU / 4 GB, 30 GiB disk.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":75,"includedUsageUsd":0,"limits":"From $75 to $200 a month, with 7,500 to 20,000 initial credits and a 1.5x top-up bonus. 20 concurrent sandboxes, 200 per day, largest shape 8 vCPU / 8 GB, 50 GiB disk.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"New accounts start with 500 free credits, no card required; the Free plan allows one sandbox at a time.","notes":["Rates are published per hour on the sandbox pricing page; the billing docs give the same rates per second ($0.0000100455 per vCPU-second, $0.0000032195 per GB-second).","Disk storage is $0.0001584 per GB-hour (x 730 = $0.115632 per GB-month, the storage rate we show) and is billed whether the sandbox runs or is paused.","Egress is unmetered; each sandbox has a 50 GiB bandwidth budget by default that can be recharged.","Plan fees are ranges ($10-$50 Beginner, $75-$200 Pro); we show the lowest fee. The sandbox pages do not give a dollar value for the included credits, so we show no included usage.","Concurrent caps count paused sandboxes too; only destroying a sandbox frees a slot. Sandboxes per day: 10 / 50 / 200 / 300 on Free / Beginner / Pro / Enterprise.","No maximum session length: a sandbox runs until destroyed or paused. Auto-pause on idle is off by default (60 to 86,400 seconds when set).","Enterprise allows 8 vCPU / 16 GB and 30 concurrent sandboxes; the maximum size we show is the self-serve Pro limit.","Up to five snapshots per user. Outbound traffic is open until any egress rule is set, then deny-by-default."]},"bestFor":["Agents that sit idle between steps, since pause keeps full memory state and stops vCPU billing","Workloads that move a lot of data out, because egress is not charged","Multi-agent setups that need sandboxes talking over a private network"],"watchOut":["Paused sandboxes still pay for memory and disk","The Free plan runs one sandbox at a time, up to 1 vCPU / 1 GB","No Python SDK yet; Python goes through the REST API","Concurrent limits count paused sandboxes until they are destroyed"],"sources":[{"url":"https://createos.sh/pricing/sandbox","quote":"Per vCPU-hour $0.03616363","checked":"2026-11-13"},{"url":"https://createos.sh/pricing/sandbox","quote":"Per GiB-hour of memory $0.01159025","checked":"2026-11-13"},{"url":"https://createos.sh/pricing/sandbox","quote":"Paused sandboxes are billed for memory and storage only. Your state is preserved and there is no vCPU charge.","checked":"2026-11-13"},{"url":"https://createos.sh/products/sandbox","quote":"Per GB-hour, storage $0.0001584","checked":"2026-11-13"},{"url":"https://createos.sh/products/sandbox","quote":"New accounts start with 500 free credits.","checked":"2026-11-13"},{"url":"https://createos.sh/products/sandbox","quote":"CreateOS Sandbox runs each workload in a Firecracker micro-VM with its own kernel.","checked":"2026-11-13"},{"url":"https://createos.sh/docs/Sandbox/Limits/","quote":"Free is for trying the API (one sandbox at a time, no disks or templates)","checked":"2026-11-13"},{"url":"https://createos.sh/docs/Sandbox/Limits/","quote":"A sandbox runs until you destroy it, pause it, or it auto-pauses. There is no lifetime cap to fit a job inside.","checked":"2026-11-13"},{"url":"https://createos.sh/docs/Sandbox/Limits/","quote":"No GPU shapes are offered.","checked":"2026-11-13"},{"url":"https://createos.sh/docs/Account-Billing/Pricing","quote":"Choose between subscription tiers with bonus multipliers or pure top-up credits to match your usage patterns.","checked":"2026-11-13"},{"url":"https://createos.sh/llms.txt","quote":"The self-serve platform has a free tier with no card required.","checked":"2026-11-13"},{"url":"https://createos.sh/docs/Account-Billing/Pricing","quote":"$75 - $200","checked":"2026-11-13"},{"url":"https://createos.sh/docs/Sandbox/Limits/","quote":"The largest shapes (4 vCPU / 8 GB and 8 vCPU / 8 GB on Pro, 8 vCPU / 16 GB on Enterprise) unlock with the plan.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/createos","score":{"code-sandbox":{"total":72,"rank":12,"of":63}},"comparablePrice":{"code-sandbox":20.88913376},"benchmarks":[]},{"slug":"cubesandbox","name":"Tencent CubeSandbox","company":"Tencent Cloud","website":"https://github.com/TencentCloud/CubeSandbox","docsUrl":"https://github.com/TencentCloud/CubeSandbox/blob/master/docs/index.md","pricingUrl":"https://github.com/TencentCloud/CubeSandbox","categories":["code-sandbox"],"tagline":"Self-hosted KVM microVM sandbox service from Tencent that works with the E2B SDK.","summary":"CubeSandbox is an open-source sandbox service built on RustVMM and KVM that runs each sandbox in its own microVM with a dedicated kernel. It accepts the E2B SDK, so clients can switch by changing one environment variable, and supports pause and resume, snapshots, cloning and rollback, eBPF egress filtering and a credential-injecting proxy. It deploys on a single node, bare metal, cloud VMs, Kubernetes (preview) or Tencent Cloud via Terraform, and includes a web console.","isolation":"MicroVM per sandbox with a dedicated guest kernel (RustVMM on KVM), plus eBPF network isolation between sandboxes","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/TencentCloud/CubeSandbox","lastRelease":"2026-11-12","sdks":["E2B SDK (Python, JavaScript/TypeScript)","REST API","Web console"],"regions":"Wherever you deploy it: x86_64 Linux with KVM on bare metal or cloud VMs, Kubernetes, or Tencent Cloud via Terraform.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own infrastructure","billedWhile":"Not applicable: you pay only for your own hardware or cloud machines","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software, licensed Apache-2.0 apart from listed third-party components.","notes":["The license file states Apache-2.0 for CubeSandbox itself, with exceptions for bundled third-party components.","GPU sandboxes are listed on the roadmap, not as a current feature.","Cross-node pause and resume and Kubernetes deployment are marked as preview."]},"bestFor":["Moving E2B SDK code to sandboxes on your own servers without changing client code","Dense deployments with many small microVM sandboxes per host","Agents that need checkpoints, rollback or forking of a running sandbox"],"watchOut":["The quick-start path needs an x86_64 Linux host with KVM","No GPU support yet; it is on the roadmap","Kubernetes deployment and cross-node resume are still in preview"],"sources":[{"url":"https://github.com/TencentCloud/CubeSandbox","quote":"Cube Sandbox is a high-performance, out-of-the-box secure sandbox service built on RustVMM and KVM.","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox","quote":"Every sandbox runs a dedicated OS kernel in its own MicroVM.","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox","quote":"Switch from E2B Cloud seamlessly by changing one environment variable","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox","quote":"Create checkpoints on running sandboxes, roll back to any saved state at any time, or fork from a specific state to explore in parallel.","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox","quote":"Attach host GPUs to sandboxes so Agent and inference workloads can run CUDA inside the isolated VM","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox","quote":"Cube Sandbox requires an x86_64 Linux environment with KVM support.","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox/blob/master/LICENSE","quote":"Cube Sandbox is licensed under the Apache-2.0 except for the third-party components listed below.","checked":"2026-11-13"},{"url":"https://github.com/TencentCloud/CubeSandbox/releases/tag/v0.8.0","quote":"v0.8.0","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/cubesandbox","score":{"code-sandbox":{"total":68,"rank":21,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"cursor-cloud-agents","name":"Cursor cloud agents","company":"Cursor","website":"https://cursor.com","docsUrl":"https://cursor.com/docs/cloud-agent","pricingUrl":"https://cursor.com/pricing","categories":["coding-agent"],"tagline":"Cursor's agents running in Firecracker microVMs in the cloud, paid from plan usage from the $30 Premium tier up.","summary":"Cursor cloud agents run in their own isolated Ubuntu VM on Firecracker-based microVM infrastructure, with your repositories, dependencies, secrets and a full desktop that the agent can control. Environments are set up by the agent, from a saved snapshot, or from a Dockerfile. The machine has no separate price: cloud agents draw from the plan's usage pool at the selected model's rate, and they need the Premium plan ($30/month) or higher. Idle VMs are hibernated and later deleted; disk snapshots are kept for 30 days of inactivity.","isolation":"Dedicated VM per agent on Firecracker-based microVM infrastructure, in an AWS account separate from Cursor's other services","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Desktop app","Web and mobile","REST API","Slack, GitHub and Linear integrations"],"regions":"Not stated.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"No separate machine billing; usage is charged at the selected model's rate","billedWhile":"Not billed by machine time. Each run draws from the plan's usage pool at the model's token rate; third-party models are billed at API list price.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Premium","monthlyFee":30,"includedUsageUsd":0,"limits":"Includes cloud agents and automations. 4x the usage of Plus in a pool shared across Cursor and Grok, refreshed weekly. Third-party models are not covered by the pool and are billed at API list price.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Super","monthlyFee":100,"includedUsageUsd":0,"limits":"15x the usage of Plus, refreshed weekly, plus priority model access at peak times. Third-party models billed separately at API list price.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Ultra","monthlyFee":200,"includedUsageUsd":0,"limits":"50x the usage of Plus, refreshed weekly, plus earliest access to products and models. Third-party models billed separately at API list price.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Teams","monthlyFee":30,"includedUsageUsd":0,"limits":"From $30 per user per month, 4-50x Plus usage depending on seat tier, centralized billing and SSO. Price shown is per user.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"None. The Free and Plus ($8/month) plans do not include cloud agents.","notes":["The cloud machine is not priced separately: cloud agents use the plan's usage like other agent work, at the selected model's rate. Larger context windows raise token usage. Cursor asks you to set a spend limit when you first use them.","Usage pools are shared across Cursor and Grok products and refresh weekly; top-offs are available when the pool runs out. Usage amounts are stated only as multiples of the Plus plan.","Each agent runs on a default VM profile with limited memory and CPU; sizes are not published. Enterprise customers can ask support for higher limits.","Idle VMs are hibernated and then deleted on lifecycle timers; the timer refreshes on follow-up prompts. Disk snapshots used to start and resume agents are deleted after 30 days of inactivity and cannot be deleted on demand.","Environments can be set up by the agent, from a saved snapshot, or from a Dockerfile in .cursor/environment.json. Docker, Tailscale and Cloudflare Tunnel can run inside the VM.","Internet access is on by default and can be restricted to allowlisted domains; Enterprise admins can lock the policy org-wide. Runtime secrets are redacted from transcripts and commits.","Self-hosted machines can run tool execution on hardware you manage. Legacy request-based plans price cloud agents at API rates plus the Max Mode upcharge."]},"bestFor":["Cursor users who want to hand tasks to the cloud from the editor, web, phone or Slack.","Repositories that need a custom Dockerfile-based environment.","Agents that must click through a desktop or browser to test their own changes."],"watchOut":["Not included on the Free or Plus plans.","Machine sizes are not published, and only Enterprise can ask for more CPU or memory.","Third-party models are billed at API list price on top of the plan fee.","Snapshots containing cloned code are kept for 30 days of inactivity and cannot be deleted on demand."],"sources":[{"url":"https://cursor.com/pricing","quote":"Cloud agents & automations","checked":"2026-11-14"},{"url":"https://cursor.com/pricing","quote":"Teams $30 +/user/month","checked":"2026-11-14"},{"url":"https://cursor.com/pricing","quote":"Your allowance refreshes weekly, so one heavy day never wipes out your whole month.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent","quote":"Cloud Agents use your plan's usage like other agent work, at the selected model's rate.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent","quote":"You need a plan that includes Cloud Agents: Premium and higher, Teams, Enterprise, or a legacy paid plan. Free and Plus don't include them.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent/security","quote":"Runtime workspaces run on Firecracker-based microVM infrastructure.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent/security","quote":"VM runtime resources are hibernated and then deleted on lifecycle timers once the run is idle.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent/security","quote":"Snapshots are deleted after 30 days of inactivity.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent/security","quote":"Internet access is on by default but can be restricted to allowlisted domains, down to allowlist-only, and locked org-wide.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent/setup","quote":"Each cloud agent runs on a default VM profile with limited memory and CPU.","checked":"2026-11-14"},{"url":"https://cursor.com/docs/cloud-agent/setup","quote":"Manually configure the environment with a Dockerfile.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/cursor-cloud-agents","score":{"coding-agent":{"total":64,"rank":5,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"daytona","name":"Daytona","company":"Daytona","website":"https://www.daytona.io","docsUrl":"https://www.daytona.io/docs/en/sandboxes","pricingUrl":"https://www.daytona.io/pricing","categories":["code-sandbox","dev-box","gpu"],"tagline":"Pay-as-you-go sandboxes (container, Linux VM, Windows, GPU) billed per second for reserved vCPU, RAM and disk.","summary":"Daytona sandboxes are isolated environments with their own kernel, filesystem and network stack, offered as container, Linux VM, Windows, macOS and GPU sandbox classes. Billing is per second for the vCPU, RAM and disk reserved while a sandbox is started; stopped or paused sandboxes are billed for disk only, and archived sandboxes are not billed. There is no monthly plan fee: organizations move through usage tiers (Tier 1 to 4) by verifying and topping up credit, which raises resource pools, rate limits and network freedom. SDKs cover Python, TypeScript, Ruby, Go and Java.","isolation":"Linux containers by default, each with its own kernel, filesystem and network stack; Linux VM and Windows sandbox classes run as full virtual machines and add pause, resume and fork.","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","Ruby","Go","Java"],"regions":"Shared regions: United States (us) and Europe (eu). GPU sandboxes run in a global 'earth' region. Dedicated regions and custom (bring-your-own-compute) regions are also offered.","features":{"pauseResume":true,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":4,"maxMemoryGib":8},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":0.07884,"billingIncrement":"per second","billedWhile":"Wall-clock for reserved resources: vCPU, RAM and disk while started (and while creating, starting, stopping or pausing); disk only while stopped or paused; nothing while archived or deleted.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"Nvidia B300 (on-demand)","perHour":6.25},{"type":"Nvidia B200 (on-demand)","perHour":6.25},{"type":"AMD MI355X (on-demand)","perHour":5.99},{"type":"Nvidia H200 (on-demand)","perHour":4.54},{"type":"Nvidia H100 (on-demand)","perHour":3.95},{"type":"Nvidia RTX PRO 6000 (on-demand)","perHour":3.03},{"type":"Nvidia L40S (on-demand)","perHour":1.99},{"type":"Nvidia RTX 5090 (on-demand)","perHour":1.29},{"type":"Nvidia RTX 4090 (on-demand)","perHour":0.99},{"type":"Nvidia B300 (preemptible)","perHour":4.08},{"type":"Nvidia B200 (preemptible)","perHour":3.59},{"type":"AMD MI355X (preemptible)","perHour":3.44},{"type":"Nvidia H200 (preemptible)","perHour":2.61},{"type":"Nvidia H100 (preemptible)","perHour":2.27},{"type":"Nvidia RTX PRO 6000 (preemptible)","perHour":1.74},{"type":"Nvidia L40S (preemptible)","perHour":1.14},{"type":"Nvidia RTX 5090 (preemptible)","perHour":0.74},{"type":"Nvidia RTX 4090 (preemptible)","perHour":0.57}],"plans":[{"name":"Pay as you go (Tier 1-4)","monthlyFee":0,"includedUsageUsd":0,"limits":"No plan fee. Org-wide pools by tier: Tier 1 (email verified) 10 vCPU / 10-20 GiB RAM (docs conflict) / 30 GiB disk; Tier 2 (card + $25 top-up) 100 / 200 / 300; Tier 3 ($500 top-up) 250 / 500 / 2,000; Tier 4 ($2,000 top-up every 30 days) 500 / 1,000 / 5,000. Per sandbox: max 4 vCPU, 8 GiB RAM, 10 GiB disk (non-GPU). Tier 1-2 cannot override the default network restrictions.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"$200 in free compute on sign-up, no credit card required; free credits cannot be used for GPU sandboxes. Startups can apply for up to $50k in credits.","notes":["CPU and memory rates are per reserved resource: $0.0504/vCPU-hour and $0.0162/GiB-hour, billed per second.","Storage is $0.000108 per GiB-hour after the first 5 GiB free; $0.07884 per GiB-month is our conversion at 730 hours/month.","Windows sandboxes are priced at $0.0858/vCPU-hour instead of $0.0504.","Stopped and paused sandboxes keep paying for reserved disk; archive container sandboxes (moves the filesystem to object storage) to stop all billing. Snapshots created from a sandbox remain billed for storage after the sandbox is deleted.","Sandboxes auto-stop after 15 minutes of inactivity by default; a running process alone does not count as activity, so long jobs can be stopped mid-run unless auto-stop is set to 0.","Pause/resume (memory preserved) is available only for Linux VM and Windows sandboxes, not container or GPU sandboxes. GPU sandboxes are ephemeral and are deleted when stopped.","Ambiguity: the pricing page says '$200 in free compute included' without saying whether it recurs; we treat it as a one-time credit.","Ambiguity: the limits page lists Tier 1 memory as 10 GiB in the Tiers table and 20 GiB in the Limits table.","No egress price is published on the pages checked."]},"bestFor":["Long-lived agent workspaces that can be stopped or archived between uses to cut cost to disk-only or zero.","Agents that need GPU sandboxes (H100, H200, B200, B300, RTX, MI355X) with on-demand or cheaper preemptible pricing.","Teams wanting SDKs beyond Python/TypeScript (Ruby, Go, Java) or Windows sandboxes."],"watchOut":["Default per-sandbox size limit is 4 vCPU / 8 GiB RAM / 10 GiB disk for non-GPU sandboxes.","Tier 1 and Tier 2 organizations cannot change network restrictions per sandbox; full internet access needs Tier 3 ($500 top-up).","Billing is for reserved resources, not usage: an idle started sandbox pays the full vCPU and RAM rate.","The public GitHub repository stopped being maintained in June 2026; core development moved to a private codebase."],"sources":[{"url":"https://www.daytona.io/pricing","quote":"Compute vCPU $0.0504/h Memory GiB $0.0162/h Storage GiB $0.000108/h * Price per GiB after first 5 free","checked":"2026-11-14"},{"url":"https://www.daytona.io/pricing","quote":"OS Windows $0.0858/vCPU/h","checked":"2026-11-14"},{"url":"https://www.daytona.io/pricing","quote":"On-demand ($/hour) Nvidia B300 $6.25/h Nvidia B200 $6.25/h AMD MI355X $5.99/h Nvidia H200 $4.54/h Nvidia H100 $3.95/h","checked":"2026-11-14"},{"url":"https://www.daytona.io/pricing","quote":"Nvidia RTX PRO 6000 $3.03/h Nvidia L40S $1.99/h Nvidia RTX 5090 $1.29/h Nvidia RTX 4090 $0.99/h","checked":"2026-11-14"},{"url":"https://www.daytona.io/pricing","quote":"Preemptible ($/hour) Nvidia B300 $4.08/h Nvidia B200 $3.59/h AMD MI355X $3.44/h Nvidia H200 $2.61/h Nvidia H100 $2.27/h Nvidia RTX PRO 6000 $1.74/h Nvidia L40S $1.14/h Nvidia RTX 5090 $0.74/h Nvidia RTX 4090 $0.57/h","checked":"2026-11-14"},{"url":"https://www.daytona.io/pricing","quote":"$200 in free compute included.","checked":"2026-11-14"},{"url":"https://www.daytona.io/pricing","quote":"All billing is calculated per second.","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/billing","quote":"Sandboxes are billed for the resources reserved: vCPU , RAM , and disk , depending on the sandbox lifecycle state","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/billing","quote":"Free credit balance : the amount of free credits; not available for GPU sandboxes","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/limits","quote":"Credit card linked, $25 top-up","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/sandboxes","quote":"Organizations get a maximum sandbox resource limit of 4 vCPUs , 8GB RAM , and 10GB disk .","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/sandboxes","quote":"if not set, the default interval of 15 minutes is used","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/network-limits","quote":"Tier 1 & Tier 2 : Network access is restricted and cannot be overridden at the sandbox level.","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/regions","quote":"Regions managed by Daytona and available to all organizations","checked":"2026-11-14"},{"url":"https://github.com/daytonaio/daytona","quote":"As of June 2026, Daytona's core development has moved to a private codebase.","checked":"2026-11-14"},{"url":"https://www.daytona.io/docs/en/sandboxes","quote":"Daytona also provides VM sandboxes with a dedicated Linux VM or Windows operating system","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/daytona","score":{"code-sandbox":{"total":64,"rank":30,"of":63},"dev-box":{"total":65,"rank":17,"of":38},"gpu":{"total":71,"rank":3,"of":7}},"comparablePrice":{"code-sandbox":29.1456,"dev-box":36.432,"gpu":2.27},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":682,"unit":"ms","better":"lower","date":"2026-11-13","rank":12,"of":27},{"source":"aisandboxbench","metric":"Cold start (median)","value":1.25,"unit":"s","better":"lower","date":"2026-10-23","rank":8,"of":9},{"source":"aisandboxbench","metric":"Agent tool-call replay (median)","value":12.274,"unit":"s","better":"lower","date":"2026-10-23","rank":1,"of":9},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":21.52,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":3,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":11.8,"unit":"s","better":"lower","date":"2026-09-29","rank":2,"of":11}]},{"slug":"depot","name":"Depot","company":"Depot","website":"https://depot.dev","docsUrl":"https://depot.dev/docs","pricingUrl":"https://depot.dev/pricing","categories":["ci-runner","code-sandbox"],"tagline":"CI runners, container builds and microVM sandboxes, billed per second at $0.18 per vCPU-hour with memory included.","summary":"Depot sells fast CI: its own Depot CI, drop-in GitHub Actions runners and remote container builds. The same compute also powers Depot Sandboxes, Linux microVMs created from a Node.js SDK, and Depot Managed Agents. CI sandboxes and GitHub Actions runners cost $0.0001 per second for 2 vCPU and 8 GB, scaling linearly to 64 vCPU and 256 GB. Plans start at $20 a month with included minutes.","isolation":"Depot CI and Sandboxes run in microVMs on Depot Metal; GitHub Actions runners use ephemeral single-tenant EC2 instances","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Node.js (@depot/sandbox)","CLI","API","GitHub Actions runner labels"],"regions":"US and EU; custom regions on the Business plan.","features":{"pauseResume":null,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":64,"maxMemoryGib":256},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.36,"instance":"2 vCPU / 8 GB (base CI sandbox or GitHub Actions runner)","storageGibMonth":null,"billingIncrement":"per second (Depot CI and Sandboxes); per minute (GitHub Actions runners)","billedWhile":"While the job or sandbox runs, by allocated vCPU; memory is included in the vCPU price.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Developer","monthlyFee":20,"includedUsageUsd":0,"limits":"1 user. Each month: 2,000 sandbox minutes (shared by Depot CI, Sandboxes and Managed Agents, measured at 2 vCPUs), 2,000 GitHub Actions minutes, 500 container build minutes, 25 GB cache and registry storage.","includedVcpuHours":66.6667,"includedGibHours":0,"capped":false},{"name":"Startup","monthlyFee":200,"includedUsageUsd":0,"limits":"Unlimited users. Each month: 20,000 sandbox minutes (shared, measured at 2 vCPUs), 20,000 GitHub Actions minutes, 5,000 container build minutes, 250 GB cache and registry storage.","includedVcpuHours":666.6667,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"7-day free trial, no credit card required.","notes":["Sandbox compute (Depot CI, Sandboxes, Managed Agents) is $0.00005 per second per vCPU, which is $0.18 per vCPU-hour; the base 2 vCPU / 8 GB size is $0.0001 per second ($0.36/hour). GitHub Actions runners are $0.006 per minute for 2 vCPU / 8 GB, the same hourly rate.","Sizes scale linearly from 2 vCPU / 8 GB to 64 vCPU / 256 GB on Intel and Arm. Depot writes GB, which we treat as GiB.","Included sandbox minutes are measured at 2 vCPUs, so 2,000 minutes equal about 66.7 vCPU-hours. GitHub Actions minutes and container build minutes are separate allowances not counted here.","Sandboxes default to 2 vCPU, 4,096 MB memory and 100 GB disk, accept 2 to 64 vCPUs, and live at most 24 hours (default 120 minutes).","Sandbox snapshots save files in the home directory and /workspace (up to 25 GiB), not running processes or memory.","Container builds cost $0.04 per minute on the default 16-vCPU builder. Cache and registry storage is $0.20 per GB-month beyond the included amount.","Static IPs, firewall rules, GPU builds and running in your own AWS account are Business-plan features."]},"bestFor":["Speeding up GitHub Actions with a one-line runner label change","Teams that want CI and agent sandboxes on one bill and one allowance","Large CI jobs that need up to 64 vCPU and 256 GB"],"watchOut":["No free plan after the 7-day trial; paid plans start at $20 a month","Sandbox SDK is Node.js only and custom images must be Depot snapshot images","Sandboxes live at most 24 hours, and snapshots keep files only","Billed on allocated vCPU for the whole run, not on CPU actually used"],"sources":[{"url":"https://depot.dev/pricing","quote":"$0.00005 / second / vCPU Included minutes are measured at 2 vCPUs","checked":"2026-11-13"},{"url":"https://depot.dev/pricing","quote":"$0.006 / minute Base 2-vCPU Linux runner","checked":"2026-11-13"},{"url":"https://depot.dev/pricing","quote":"2,000 sandbox minutes","checked":"2026-11-13"},{"url":"https://depot.dev/pricing","quote":"20,000 sandbox minutes","checked":"2026-11-13"},{"url":"https://depot.dev/pricing","quote":"|Sandbox sizes |2 to 64 vCPUs, Intel or Arm |","checked":"2026-11-13"},{"url":"https://depot.dev/pricing","quote":"|US & EU regions |Included |Included |Included |","checked":"2026-11-13"},{"url":"https://depot.dev/docs/ci/overview","quote":"| `depot-ubuntu-24.04`    | `2x8`        | 2    | 8 GB   | $0.0001          | 1x                      |","checked":"2026-11-13"},{"url":"https://depot.dev/docs/ci/overview","quote":"Sandboxes can use from 2 CPUs/8 GB of memory up to 64 CPUs/256 GB of memory.","checked":"2026-11-13"},{"url":"https://depot.dev/docs/github-actions/runner-types","quote":"| `depot-ubuntu-24.04`    | 2    | 8 GB   | 100 GB    | 2GB                   | $0.006           | 1x                 |","checked":"2026-11-13"},{"url":"https://depot.dev/docs/sandboxes/overview","quote":"Depot Sandboxes give you isolated Linux microVMs for running code, tests, and agents on x86-64 or ARM64.","checked":"2026-11-13"},{"url":"https://depot.dev/docs/api/sandbox-sdk-reference","quote":"The maximum sandbox lifetime is 24 hours.","checked":"2026-11-13"},{"url":"https://depot.dev/docs/api/sandbox-sdk-reference","quote":"Snapshots preserve files, not running processes or memory, and are limited to 25 GiB of file data.","checked":"2026-11-13"},{"url":"https://depot.dev/docs/github-actions/overview","quote":"launches ephemeral, single-tenant EC2 instances in response to your GitHub workflow jobs","checked":"2026-11-13"},{"url":"https://depot.dev/docs","quote":"for a Depot account to get a free trial. No credit card required.","checked":"2026-11-13"},{"url":"https://depot.dev/pricing","quote":"START A 7-DAY TRIAL","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/depot","score":{"ci-runner":{"total":58,"rank":9,"of":10},"code-sandbox":{"total":55,"rank":50,"of":63}},"comparablePrice":{"ci-runner":56,"code-sandbox":83.36},"benchmarks":[]},{"slug":"devin","name":"Devin","company":"Cognition","website":"https://devin.ai","docsUrl":"https://docs.devin.ai","pricingUrl":"https://devin.ai/pricing","categories":["coding-agent"],"tagline":"Autonomous coding agent whose cloud sessions run in their own VM; machine time is folded into plan usage.","summary":"Devin Cloud runs each session on its own machine, booted from a snapshot of an environment you describe in a blueprint, with Linux by default and Windows or macOS VMs as options. Sessions go to sleep after 5 minutes of inactivity and use nothing while asleep. Machine time is not sold separately: it counts toward the plan's usage allowance together with the agent's actions, and Devin says VM time is typically a small fraction of usage. Paid plans get an allowance that refreshes daily and weekly; extra usage is bought as credits consumed at API pricing. Cloud agents start on the $20 Pro plan; the free plan only has a light quota for local agent use.","isolation":"Each session, including each managed sub-session, runs in its own isolated virtual machine","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Web app","CLI","REST API","Devin Desktop"],"regions":"Not stated for self-serve plans. Enterprise can deploy in its own VPC.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"No separate machine billing; the usage allowance covers agent work and VM time","billedWhile":"Usage accrues while Devin works; a few units keep the VM running while awake. Asleep sessions use nothing; sessions sleep after 5 minutes of inactivity by default (30 minutes for some Windows or GPU sessions).","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Light quota to code with agents, limited model availability, unlimited inline edits and Tab completions. Access to cloud agents (Devin Cloud) is listed only from Pro up.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":20,"includedUsageUsd":0,"limits":"Single user. Includes access to cloud agents (Devin Cloud). Usage allowance refreshes daily and weekly and is shared between cloud sessions, Devin CLI and Devin Desktop; up to 10 concurrent sessions. Extra usage past the allowance is bought as credits consumed at API pricing.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":200,"includedUsageUsd":0,"limits":"Single user. Significantly higher quota, refreshed weekly with no daily cap; up to 10 concurrent sessions; extra usage at API pricing.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Teams","monthlyFee":80,"includedUsageUsd":0,"limits":"$80 a month for the team plan plus $40 a month per full seat, as listed on the pricing page. Full seats get a Pro-sized allowance; free flex seats draw from shared extra-usage credits. Up to 100 users, unlimited concurrent sessions, plus 2 concurrent sessions per user in Free mode (SWE-2) at no cost.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"None for cloud sessions. The $0 Free plan has a light quota for local agent use; access to cloud agents (Devin Cloud) starts with Pro.","notes":["The cloud machine is not priced separately. Usage covers the number and complexity of the agent's actions plus virtual machine time, which Devin describes as typically a small fraction of total usage. No per-hour machine rate is published.","Each paid plan has a usage allowance that refreshes automatically: daily and weekly on Pro and Teams full seats, weekly only on Max. Allowance sizes are not published in numbers. Past the allowance, extra usage is bought as prepaid credits consumed at API pricing; purchased credits roll over and do not expire.","Ambiguity on Teams pricing: the pricing page lists $80 a month for the team plan plus $40 a month per full seat, while Devin's billing docs describe $80 as a monthly minimum that full seats ($40 each) count toward, with any shortfall charged as prepaid credits. We show the $80 base.","Devin uses nothing while waiting for your reply, waiting for a test suite, or setting up and cloning repositories, apart from the few units needed to keep the VM running.","Windows sessions use about 9% more usage than Linux sessions; macOS sessions currently cost the same as Linux (promotional launch pricing).","Every session boots from the organization's environment snapshot, rebuilt automatically when the configuration changes. Session changes do not persist back to the snapshot.","Machine sizes, memory and session time limits are not published for the hosted VMs.","Network allowlists are set through security profiles and enforced on Devin-managed VMs. Enterprise has custom pricing based on Agent Compute Units, can deploy in its own VPC, and can run sessions on its own infrastructure.","Pro, Max and Teams accounts can link a ChatGPT Go, Plus or Pro subscription so GPT model usage is billed there instead of the Devin allowance."]},"bestFor":["Handing off whole tickets that need a full machine, browser and desktop to test changes.","Teams that want shared environment snapshots built from a declarative blueprint.","Projects that need Windows or macOS build machines as well as Linux."],"watchOut":["No published machine sizes or per-hour machine price.","Allowance sizes are not stated in numbers; past the allowance you pay for extra usage at API pricing.","Pro and Max allow at most 10 concurrent sessions.","The Free plan has no cloud agents.","The pricing page and billing docs describe the Teams $80 fee differently."],"sources":[{"url":"https://devin.ai/pricing","quote":"Light quota to code with agents","checked":"2026-11-14"},{"url":"https://devin.ai/pricing","quote":"Access to cloud agents (Devin Cloud)","checked":"2026-11-14"},{"url":"https://devin.ai/pricing","quote":"$80/month for team plan + $40/mo per full dev seat","checked":"2026-11-14"},{"url":"https://devin.ai/pricing","quote":"2 concurrent sessions in Free mode per user","checked":"2026-11-14"},{"url":"https://devin.ai/pricing","quote":"Each paid plan comes with a usage allowance that refreshes automatically on a daily and weekly basis.","checked":"2026-11-14"},{"url":"https://devin.ai/pricing","quote":"If you do go beyond your included usage on a paid plan, you can purchase extra usage which is consumed at API pricing.","checked":"2026-11-14"},{"url":"https://devin.ai/pricing","quote":"Deploy in your virtual private cloud (VPC)","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/self-serve","quote":"Pro | Individual users | $20/month | 1","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/self-serve","quote":"Max | Power users who need more quota | $200/month | 1","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/self-serve","quote":"Teams | Teams working with Devin together | $80/month minimum | Up to 100","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/self-serve","quote":"When you have fewer than two full seats, the remainder of the $80 minimum is automatically charged as prepaid on-demand credits that the whole team can draw from.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/self-serve","quote":"Each member can run up to 2 Free sessions at a time","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/self-serve","quote":"Purchased credits never expire.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/usage","quote":"Virtual machine time (typically a small fraction of total usage)","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/usage","quote":"Devin sleeps automatically after 5 minutes of inactivity by default.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/usage","quote":"Some sessions, such as those on Windows or GPU machines, may sleep after 30 minutes of inactivity instead.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/admin/billing/usage","quote":"Windows sessions consume approximately 9% more usage than equivalent Linux (Ubuntu) sessions.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/release-notes/2026","quote":"Each managed Devin is a full Devin with its own isolated virtual machine.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/onboard-devin/environment","quote":"Every session boots from a snapshot, a frozen, bootable image of the environment.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/onboard-devin/environment","quote":"When your configuration changes, a new snapshot is built automatically.","checked":"2026-11-14"},{"url":"https://docs.devin.ai/product-guides/security-profiles","quote":"A network policy is an allowlist of destinations the session's machine is permitted to reach.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/devin","score":{"coding-agent":{"total":64,"rank":6,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"digitalocean-harness-runtime","name":"DigitalOcean Agent Harness Runtime","company":"DigitalOcean","website":"https://docs.digitalocean.com/products/managed-agents/","docsUrl":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/","pricingUrl":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","categories":["code-sandbox"],"tagline":"Hardware-isolated microVM sandboxes for agents and code, billed per second on CPU used and memory allocated.","summary":"Harness Runtime is part of DigitalOcean Managed Agents, in public preview for all users. Each session runs in a hardware-isolated microVM that can host a coding agent such as Claude Code or Codex, or act as a plain sandbox for running commands. CPU is billed on actual consumption and memory on allocation, per second, and paused sessions stop compute charges while keeping files and history. Checkpoints capture the whole machine so you can fork or roll back.","isolation":"Hardware-isolated microVM per session","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["doctl CLI","REST API","PyDo (Python)","Control panel","Cuttlefish iOS app"],"regions":"RIC1 (Richmond) only, per the availability table.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.044,"gibHour":0.0095,"instanceHour":null,"instance":null,"storageGibMonth":0.05,"billingIncrement":"per second","billedWhile":"While a session is active: CPU on actual consumption (zero while the agent waits), memory on the allocated size over time. Paused sessions pay no compute; volumes, checkpoints and snapshots keep billing as storage.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Agent Droplets Pro","monthlyFee":50,"includedUsageUsd":58.82,"limits":"$50/month fee funds a balance spent at a 15% discount, covering $58.82 of eligible usage at standard rates. Usage beyond that bills at standard rates. Unlimited agents subject to fair use.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Agent Droplets Team","monthlyFee":200,"includedUsageUsd":250,"limits":"$200/month fee funds a balance spent at a 20% discount, covering $250 of eligible usage at standard rates. Usage beyond that bills at standard rates. Unlimited agents subject to fair use.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"New customers receive a one-time $5 promotional credit for Managed Agents usage.","notes":["Harness Runtime and the Agent Droplets plans are in public preview.","Memory is priced per GB-hour; DigitalOcean writes GB, which we treat as GiB.","Fixed sandbox sizes: 1 vCPU/1 GB, 2 vCPU/2 GB, 2 vCPU/4 GB (default), 4 vCPU/8 GB and 16 vCPU/32 GB, with 50 to 500 GB of disk. Full use of the default size costs $0.126/hour.","Session volumes, checkpoints, snapshots and custom templates are each $0.05 per GiB-month. Public internet egress is $0.01 per GiB.","Persistent workspaces are free during their public preview.","Plan discounts do not apply to third-party models such as Claude or GPT; inference on DigitalOcean-hosted models is billed separately at Inference rates.","Each session has a monthly active compute cap of 744 hours. Idle suspend defaults to 15 minutes. Up to 100 concurrent sessions per team depending on account tier.","Sessions pause when all balances are empty; prepaid balance or a plan is needed before starting sessions."]},"bestFor":["Running hosted coding agents such as Claude Code, Codex or OpenCode in isolated machines","Agent work that waits a lot on models or people, since idle CPU is not billed and sessions can pause","Branching work from a checkpoint to try several approaches in parallel"],"watchOut":["Public preview, and only one region (RIC1) is listed","Memory is billed on the allocated size even while the agent is idle","No private VPC access, live app preview or lifecycle webhooks yet","Requires prepaid funds or a plan before sessions can start"],"sources":[{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"| CPU | $0.044 per vCPU-hour | Actual CPU consumed |","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"| Memory | $0.0095 per GB-hour | Actual memory allocated over time |","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"Snapshot and checkpoint storage is billed at $0.05 per GiB-month for the data you store.","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"| Pro | $50 | 15% | $58.82 |","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"| Team | $200 | 20% | $250.00 |","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"Outbound traffic from the sandbox to the public internet is billed at $0.01 per GiB of egress data.","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/pricing/","quote":"Pausing a session stops compute charges while preserving the sandbox’s files and conversation history","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/how-to/pay-for-harness-runtime/","quote":"New customers receive a $5 promotional credit for Managed Agents usage.","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/features/","quote":"| XLarge | `mars-16vcpu-32gb` | 16 | 32 GB | 500 GB | 512 Mbps |","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/features/","quote":"You can also create checkpoints to capture the whole machine state from which you can fork a new session or roll back to a prior state.","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/limits/","quote":"Sandbox egress is unrestricted unless the environment spec sets an allowlist.","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/limits/","quote":"Each session has a monthly active compute cap of 744 hours.","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/","quote":"Managed Agents, available in public preview for all users, bring together hardware-isolated microVM sandboxes","checked":"2026-11-13"},{"url":"https://docs.digitalocean.com/products/managed-agents/agent-harness-runtime/details/availability/","quote":"Harness Runtime is available in the regions shown below.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/digitalocean-harness-runtime","score":{"code-sandbox":{"total":61,"rank":38,"of":63}},"comparablePrice":{"code-sandbox":50},"benchmarks":[]},{"slug":"docker-sandboxes","name":"Docker Sandboxes","company":"Docker","website":"https://docs.docker.com/ai/sandboxes/","docsUrl":"https://docs.docker.com/ai/sandboxes/","pricingUrl":"https://docs.docker.com/ai/sandboxes/faq/","categories":["code-sandbox"],"tagline":"Docker's sbx CLI runs coding agents in local microVMs for free, with optional pay-as-you-go cloud sandboxes.","summary":"Docker Sandboxes run AI coding agents in microVMs on your own machine, each with its own Docker daemon, created and managed with the sbx CLI. All outbound traffic passes through a host proxy that enforces network policy and injects credentials so the agent never sees real keys. The CLI and local compute are free, including for commercial use, but the software is not open source and requires a Docker sign-in. Cloud sandboxes on Docker-managed compute are billed per second through a separate pay-as-you-go subscription whose rates are not published in the docs.","isolation":"MicroVM per sandbox (hypervisor isolation) with its own Docker daemon","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":"2026-11-11","sdks":["CLI (sbx)","Sandboxes API and SDK (cloud)"],"regions":"Local sandboxes run on your machine: macOS 14+ on Apple Silicon, Windows 11 x64 with Windows Hypervisor Platform, or Ubuntu 24.04+ with KVM. Cloud sandbox regions not stated.","features":{"pauseResume":true,"snapshots":null,"gpu":true,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Local: not applicable. Cloud: metered per second (rates not published)","billedWhile":"Local: free. Cloud: by sandbox runtime and the vCPU and memory configuration","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"The sbx CLI and local sandbox compute are free, including for commercial work. Cloud sign-ups get a one-time promotional credit (amount not stated).","notes":["Not open source; the CLI is distributed as binaries and needs a Docker account sign-in.","Cloud sandboxes use a pay-as-you-go plan with no recurring fee, metered in seconds by runtime and vCPU plus memory; the per-unit rates are shown only inside the billing console, not in public docs.","Organization-wide policy management and audit logs for local sandboxes require a separate paid subscription with prices from sales.","Stopping a sandbox keeps the VM, installed packages, images and files; it persists until removed."]},"bestFor":["Running Claude Code, Codex and other coding agents on a laptop with VM isolation and full Docker access","Keeping API credentials outside the agent's environment through proxy injection","Teams already standardized on Docker tooling"],"watchOut":["Not open source and requires a Docker sign-in","Cloud sandbox prices are not published in the docs","Each sandbox runs a full VM and its own Docker daemon, so overhead is higher than a container"],"sources":[{"url":"https://docs.docker.com/ai/sandboxes/","quote":"The sbx CLI and local sandbox compute are free to use, including for commercial work.","checked":"2026-11-13"},{"url":"https://docs.docker.com/ai/sandboxes/","quote":"turn on GPU passthrough","checked":"2026-11-13"},{"url":"https://docs.docker.com/ai/sandboxes/architecture/","quote":"Sandboxes (microVMs) | Full (hypervisor) | Isolated daemon | Autonomous agents","checked":"2026-11-13"},{"url":"https://docs.docker.com/ai/sandboxes/architecture/","quote":"All outbound TCP traffic from the sandbox routes through a proxy on your host.","checked":"2026-11-13"},{"url":"https://docs.docker.com/ai/sandboxes/architecture/","quote":"You can stop and restart without recreating the VM, preserving installed packages, Docker images, and in-sandbox files.","checked":"2026-11-13"},{"url":"https://docs.docker.com/ai/sandboxes/install/","quote":"Hypervisor and KVM setup is required only to run local sandboxes.","checked":"2026-11-13"},{"url":"https://docs.docker.com/agentic-platform/signup/","quote":"Cloud sandboxes use pay-as-you-go compute with no recurring subscription fee. Docker meters compute in seconds, based on sandbox runtime and the CPU and memory configuration.","checked":"2026-11-13"},{"url":"https://docs.docker.com/subscription-billing/plans/docker-agentic-platform/","quote":"Docker Agentic Platform signups receive a one-time promotional credit toward cloud compute usage.","checked":"2026-11-13"},{"url":"https://github.com/docker/sbx-releases/releases/tag/v0.53.0","quote":"v0.53.0","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/docker-sandboxes","score":{"code-sandbox":{"total":62,"rank":33,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"e2b-runtime","name":"E2B open-source runtime","company":"E2B","website":"https://github.com/e2b-dev/runtime","docsUrl":"https://github.com/e2b-dev/runtime","pricingUrl":"https://github.com/e2b-dev/runtime","categories":["code-sandbox"],"tagline":"The Apache-2.0 backend behind E2B Cloud: Firecracker microVM sandboxes you can run on your own Linux host or cloud.","summary":"E2B Runtime is the full backend that powers E2B Cloud: the control-plane API, the per-node orchestrator that drives Firecracker, the in-VM agent, the edge router and the template builder. Sandboxes are restored from snapshots, can be paused, resumed and forked, and get per-sandbox egress allow and deny lists. The single-machine Embed package runs the whole stack on one Linux host with KVM via Docker Compose, and also ships as Terraform for GCP, AWS and Azure and as a Kubernetes manifest. The software is free; multi-node production deployments in your own cloud are offered by E2B as an enterprise service without a published price.","isolation":"One Firecracker microVM per sandbox, in its own cgroup and network namespace","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/e2b-dev/runtime","lastRelease":"2026-09-10","sdks":["JavaScript/TypeScript","Python","CLI","REST API"],"regions":"Wherever you deploy it: one Linux host with KVM, or GCP, AWS or Azure via Terraform, or Kubernetes.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own infrastructure","billedWhile":"Not applicable: you pay only for your own hardware or cloud resources","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license. E2B's hosted cloud and dedicated enterprise deployments are separate offerings.","notes":["Embed is single-machine by design; for more than one node the project points to E2B's dedicated enterprise deployment, which has no published price.","Requires Linux with KVM.","Releases are cut per package; the date shown is the most recent GitHub release."]},"bestFor":["Teams that want the same sandbox stack locally or on their own servers that E2B runs in its cloud","Agent workloads that need pause, resume and forking of running microVMs","Existing E2B SDK users who need sandboxes inside their own account"],"watchOut":["The free Embed package runs on a single machine; multi-node setups are an enterprise offering","Needs Linux hosts with KVM","Several backing services to operate: PostgreSQL, Redis, ClickHouse and object storage"],"sources":[{"url":"https://github.com/e2b-dev/runtime","quote":"E2B Runtime is the complete backend that powers E2B Cloud","checked":"2026-11-13"},{"url":"https://github.com/e2b-dev/runtime","quote":"One Firecracker microVM per sandbox, in its own cgroup and network namespace, with a per-sandbox nftables egress firewall","checked":"2026-11-13"},{"url":"https://github.com/e2b-dev/runtime","quote":"Idle sandboxes auto-pause, and incoming traffic wakes them transparently.","checked":"2026-11-13"},{"url":"https://github.com/e2b-dev/runtime","quote":"E2B Embed is the whole stack, real Firecracker sandboxes included, on a single Linux host with KVM.","checked":"2026-11-13"},{"url":"https://github.com/e2b-dev/runtime","quote":"Embed is one machine by design; when one node is not enough, see below.","checked":"2026-11-13"},{"url":"https://github.com/e2b-dev/runtime","quote":"It is written in Go, licensed under Apache-2.0","checked":"2026-11-13"},{"url":"https://github.com/e2b-dev/runtime/releases/tag/2026.30","quote":"2026.30","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/e2b-runtime","score":{"code-sandbox":{"total":68,"rank":23,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"e2b","name":"E2B","company":"E2B","website":"https://e2b.dev","docsUrl":"https://e2b.dev/docs","pricingUrl":"https://e2b.dev/pricing","categories":["code-sandbox"],"tagline":"Firecracker microVM sandboxes for AI agents, billed per second for vCPU and RAM while running.","summary":"E2B runs each sandbox as a Firecracker microVM started from a template, controlled through JavaScript/TypeScript and Python SDKs. Running sandboxes are billed per second for allocated vCPU and RAM; disk is included and paused sandboxes do not incur compute charges. A paused sandbox keeps its filesystem and memory indefinitely and resets the continuous-runtime limit (1 hour on Hobby, 24 hours on Pro). The runtime is open source under Apache-2.0, and Enterprise customers can deploy into their own AWS, Google Cloud or Azure account (BYOC).","isolation":"Firecracker microVM per sandbox","isolationKind":"microvm","openSource":true,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","Python"],"regions":"Shared E2B cloud; a shared EU cluster is available on every plan on request for workspaces with a payment method; BYOC on AWS, Google Cloud and Azure for Enterprise.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":8,"maxMemoryGib":8},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per second","billedWhile":"Wall-clock while the sandbox is running, for allocated vCPU and RAM; compute billing stops when the sandbox is paused, killed or times out.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Hobby","monthlyFee":0,"includedUsageUsd":0,"limits":"$100 one-time usage credit (not monthly). Sessions up to 1 hour of continuous runtime, 10 concurrent sandboxes, up to 8 vCPU and 8 GiB RAM, 10 GiB disk, 1 sandbox created per second.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":150,"includedUsageUsd":0,"limits":"No usage credit included. Sessions up to 24 hours of continuous runtime, 100 concurrent sandboxes (600 with a +$500/month add-on, 1,100 with a +$1,000/month add-on), 8+ vCPU and 8+ GiB RAM, 20+ GiB disk, 5 sandboxes created per second.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Hobby plan: $0/month with a one-time $100 usage credit, no credit card required; 1-hour sessions and 10 concurrent sandboxes.","notes":["Rates are the same on Hobby and Pro: $0.000014 per vCPU-second ($0.0504/vCPU-hour) and $0.0000045 per GiB-second of RAM ($0.0162/GiB-hour).","Default sandbox size on the pricing page is 2 vCPU and 4 GiB RAM ($0.000028/s + $0.000018/s = $0.1656/hour); the billing docs' cost tips mention a default of 2 vCPU and 512 MiB RAM, so the actual default RAM depends on the template.","Disk is included at no charge (10 GiB on Hobby, 20 GiB on Pro); no separate storage or egress rate is published on the pricing page.","Paused sandboxes are kept indefinitely with no compute charge; the pages checked do not state any storage charge for paused sandboxes.","Pro concurrency add-ons: 600 concurrent for +$500/month, 1,100 concurrent for +$1,000/month. Enterprise starts at a $2,000/month commit with 500 concurrent sandboxes included.","Usage is invoiced at the start of each month for the previous month; projects are blocked when credits run out and no payment method is on file."]},"bestFor":["Agent sessions that pause between turns: pausing keeps filesystem and memory and stops compute billing.","Teams that want an open-source (Apache-2.0) runtime they can inspect or run themselves.","Workloads that need egress allow/deny lists by domain or CIDR, changeable on a running sandbox."],"watchOut":["No GPU support.","Pro costs $150/month and includes no usage credit.","Hobby sandboxes stop after 1 hour of continuous runtime unless paused and resumed.","Per-sandbox size tops out at 8 vCPU / 8 GiB on Hobby; larger sizes require Pro or contacting E2B."],"sources":[{"url":"https://e2b.dev/pricing","quote":"We charge per second of a running sandbox.","checked":"2026-11-14"},{"url":"https://e2b.dev/pricing","quote":"$0.000014/s","checked":"2026-11-14"},{"url":"https://e2b.dev/pricing","quote":"$0.0000045/s","checked":"2026-11-14"},{"url":"https://e2b.dev/pricing","quote":"$100 one-time usage credit","checked":"2026-11-14"},{"url":"https://e2b.dev/pricing","quote":"100 concurrent sandboxes, expandable to 1,100","checked":"2026-11-14"},{"url":"https://e2b.dev/pricing","quote":"Enterprise starts at a $2,000/mo commit, 500 concurrent included.","checked":"2026-11-14"},{"url":"https://e2b.dev/docs/billing","quote":"Upgrading to the Pro tier does not grant additional credits.","checked":"2026-11-14"},{"url":"https://e2b.dev/docs/billing","quote":"Max continuous runtime is the time a sandbox can run without being paused","checked":"2026-11-14"},{"url":"https://e2b.dev/docs/billing","quote":"Once a sandbox is paused, killed or times out, compute billing stops immediately.","checked":"2026-11-14"},{"url":"https://e2b.dev/docs/sandbox/persistence","quote":"A paused sandbox is kept indefinitely.","checked":"2026-11-14"},{"url":"https://e2b.dev/docs/sandbox/internet-access","quote":"You can specify IP addresses, CIDR blocks, or domain names that the sandbox is allowed to use","checked":"2026-11-14"},{"url":"https://e2b.dev/docs/faq","quote":"No. E2B sandboxes are CPU-only, sized by vCPU and RAM only, with no GPU option","checked":"2026-11-14"},{"url":"https://github.com/e2b-dev/infra","quote":"One Firecracker microVM per sandbox, in its own cgroup and network namespace","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/e2b","score":{"code-sandbox":{"total":68,"rank":22,"of":63}},"comparablePrice":{"code-sandbox":29.1456},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":1280,"unit":"ms","better":"lower","date":"2026-11-13","rank":18,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":64219,"unit":"ms","better":"lower","date":"2026-11-13","rank":15,"of":25},{"source":"aisandboxbench","metric":"Cold start (median)","value":0.793,"unit":"s","better":"lower","date":"2026-10-23","rank":4,"of":9},{"source":"aisandboxbench","metric":"Agent tool-call replay (median)","value":15.468,"unit":"s","better":"lower","date":"2026-10-23","rank":3,"of":9},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":12.61,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":11,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":20.2,"unit":"s","better":"lower","date":"2026-09-29","rank":8,"of":11}]},{"slug":"exe-dev","name":"exe.dev","company":"exe.dev","website":"https://exe.dev","docsUrl":"https://exe.dev/docs","pricingUrl":"https://exe.dev/pricing","categories":["dev-box"],"tagline":"Persistent KVM VMs on local NVMe, reached over SSH and HTTPS, sold as a monthly CPU/RAM pool or metered sandbox VMs.","summary":"exe.dev gives you persistent Linux VMs with root access, reachable with ssh exe.dev and published over HTTPS with built-in login, custom domains and secret-injecting integrations. The main product is a pool: you buy a fixed amount of vCPU and memory and run as many VMs in it as you like, from $15 a month for 2 vCPU / 4 GiB on Personal. Standalone sandbox VMs with their own reserved capacity are metered per second at $0.105 per 2 vCPU / 4 GiB hour on top of a plan. Disks sit on local NVMe and are copied to another server in the region for durability.","isolation":"KVM-isolated VM per machine, each with its own VMM process","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["SSH CLI","HTTPS API","MCP server"],"regions":"Oregon, Los Angeles, New York, Dallas, Frankfurt, Tokyo, Sydney, Singapore and London (Oregon closed to new accounts). Personal pools use the account's region; Work pools can go in any region.","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.105,"instance":"Standalone VM unit: 2 vCPU, 4 GiB","storageGibMonth":0.08,"billingIncrement":"Per second, no minimum, no rounding up to the hour","billedWhile":"Standalone VMs: reserved vCPU and memory from creation until deletion, except while stopped (a stopped VM keeps its disk). Pools: billed for as long as the pool exists, empty or full.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":4,"browserHour":null},"gpu":[],"plans":[{"name":"Personal","monthlyFee":15,"includedUsageUsd":0,"limits":"One pool of 2 vCPU / 4 GiB for $15 a month (4 vCPU / 8 GiB $35, 8 vCPU / 16 GiB $75, 16 vCPU / 32 GiB $155). 50 VMs, 100 GiB of filesystem usage, 200 GiB of network bandwidth. Standalone VMs are metered on top.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Work","monthlyFee":75,"includedUsageUsd":0,"limits":"Includes 2 vCPU + 4 GiB across all pools and standalone VMs; extra capacity $0.105 per 2 vCPU / 4 GiB hour. Unlimited pools in any region, 1,000 VMs per pool, 100 GiB disk per 2 vCPU, 50 GiB bandwidth per vCPU, unlimited users and SSO. Intro price $37.50 a month for the first 3 months.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Personal plan free trial for 7 days.","notes":["The hourly price we show is for standalone (sandbox) VMs: $0.105 per 2 vCPU / 4 GiB unit-hour, linear in that unit (1 vCPU / 2 GiB $0.0525, 16 vCPU / 32 GiB $0.84). It needs a plan: on Personal it is charged on top of the subscription.","Pool pricing is the usual way to use exe.dev: a Personal 2 vCPU / 4 GiB pool at $15 a month works out to $15 / 730 = about $0.0205 per hour, shared by every VM in the pool. Pool prices are not linear in size.","Pool and plan allowances are reserved capacity, not hour or dollar credits; the Work plan's included 2 vCPU + 4 GiB is applied to whatever runs at the same time.","Additional disk is $0.08 per GiB-month and additional bandwidth $0.05 per GiB.","Standalone VMs go up to 16 vCPU / 32 GiB on Personal and 32 vCPU / 64 GiB on Work, with up to 2 GiB of memory per vCPU; Work pools above 32 vCPU (up to 512) need approval.","Stopping a VM keeps its disk; whether memory is kept is not stated."]},"bestFor":["Running many small long-lived VMs and apps on one flat monthly pool","SSH-first developers and agents who want root on a persistent Linux box","Disk-heavy builds that benefit from local NVMe"],"watchOut":["Pools are billed whether VMs run or not; only standalone VMs stop billing when stopped","Standalone sandbox VMs cost extra on top of the Personal subscription","Personal has one pool in the account's region and no team sharing","Memory is capped at 2 GiB per vCPU for standalone VMs"],"sources":[{"url":"https://exe.dev/docs/standalone-vms.md","quote":"A standalone VM's reserved capacity is metered at $0.105 per 2 vCPU / 4 GiB unit-hour, the same rate as pool capacity.","checked":"2026-11-13"},{"url":"https://exe.dev/docs/standalone-vms.md","quote":"Time is measured by the second, with no minimum charge and no rounding up to the hour.","checked":"2026-11-13"},{"url":"https://exe.dev/docs/standalone-vms.md","quote":"A stopped standalone VM is not billed for its vCPU and memory. It keeps its disk, and it still counts toward your VM limits.","checked":"2026-11-13"},{"url":"https://exe.dev/pricing","quote":"2 vCPUs, 4 GiB memory, $15 a month 4 vCPUs, 8 GiB memory, $35 a month 8 vCPUs, 16 GiB memory, $75 a month 16 vCPUs, 32 GiB memory, $155 a month","checked":"2026-11-13"},{"url":"https://exe.dev/pricing","quote":"Try free for 7 days","checked":"2026-11-13"},{"url":"https://exe.dev/pricing/compare","quote":"Max size |16 vCPUs / 32 GiB memory |32 vCPUs / 64 GiB memory |Custom","checked":"2026-11-13"},{"url":"https://exe.dev/pricing/compare","quote":"Additional disk |$0.08 / GiB / month |$0.08 / GiB / month |Custom","checked":"2026-11-13"},{"url":"https://exe.dev/pricing/compare","quote":"KVM-isolated VMs |Included |Included |Included","checked":"2026-11-13"},{"url":"https://exe.dev/docs/pools.md","quote":"Work starts at $75/month and includes 2 vCPUs + 4 GiB of capacity, counted across all of the team's pools and standalone VMs together.","checked":"2026-11-13"},{"url":"https://exe.dev/docs/regions.md","quote":"The PDX region is not accepting new accounts; LAX is the nearest substitute.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/exe-dev","score":{"dev-box":{"total":61,"rank":21,"of":38}},"comparablePrice":{"dev-box":38.099999999999994},"benchmarks":[]},{"slug":"factory","name":"Factory","company":"Factory","website":"https://factory.com","docsUrl":"https://docs.factory.com","pricingUrl":"https://factory.com/pricing","categories":["coding-agent"],"tagline":"Droid coding agents on persistent Factory-hosted computers or fresh cloud environments, paid from plan usage.","summary":"Factory's Droid agents run locally through the desktop app, CLI and SDK, or in the cloud on Factory-hosted machines: persistent Droid Computers that keep files, packages and running services between sessions, or fresh instances started from a reusable Environment snapshot. Hosted computers auto-pause when idle and resume when a session targets them. There is no separate machine price: compute for cloud environments counts toward the plan's rolling usage limits along with model usage. Plans start at $20 a month; the Free plan is bring-your-own-inference.","isolation":"Factory-hosted cloud computers; the isolation technology is not stated","isolationKind":"unknown","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Desktop app","CLI","SDK","REST API"],"regions":"Not stated. Enterprise offers data residency and on-premise deployment.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"No separate machine billing; compute counts toward the plan's usage limits","billedWhile":"Not billed by machine time. Sessions consume the plan's standard credits for model usage plus compute used in cloud environments; hosted computers auto-pause when idle.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Bring your own inference: sign in with ChatGPT, GitHub Copilot or OpenRouter, or connect your own API keys. Whether hosted cloud computers are included is not stated.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":20,"includedUsageUsd":0,"limits":"Factory-managed inference with rolling 5-hour, 7-day and 30-day usage limits that also cover cloud environment compute. Desktop, CLI and SDK; all products included.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Plus","monthlyFee":100,"includedUsageUsd":0,"limits":"About 5x the usage of Pro.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":200,"includedUsageUsd":0,"limits":"About 10x the usage of Pro.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Teams","monthlyFee":60,"includedUsageUsd":0,"limits":"$60 a month per team plus $40 a month per member, up to 10 seats, each with Pro usage limits and centralized billing.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free hosted compute is stated. The Free plan is free to use with your own model keys or subscriptions, but its access to Factory-hosted cloud computers is not described.","notes":["The cloud machine is not priced separately: Factory states that sessions consume standard credits based on model usage along with compute usage for cloud environments, all within the plan's rolling limits. No per-hour machine price or machine sizes are published.","When the standard limits run out you can switch to Droid Core (a free open-weight model pool with its own limits) or buy Extra Usage as prepaid credits ($10 minimum, no expiration).","Droid Computers are persistent: installed packages, files, running services and configuration survive between sessions. Platform-managed computers auto-pause when idle and auto-resume when a session targets them.","Environments are reusable snapshots with repositories cloned and tools installed; each new instance starts from that baseline, and changes do not flow back to it.","Availability of Environments and managed computers depends on the organization's rollout and compute settings, and some setup needs the Manager or Owner role.","Platform-managed computers block all inbound ports except the daemon port by default. Network policy controls are listed for Business plans. You can also connect your own machine with Remote Control or run sandboxes in your own cloud with OpenCompute.","Business and Enterprise plans have custom pricing and limits."]},"bestFor":["Developers who want persistent cloud computers for an agent that keep state between sessions.","Teams that run many parallel agents from a shared environment snapshot.","Mixing local, own-server and Factory-hosted machines in one agent workflow."],"watchOut":["Machine sizes, isolation and limits of hosted computers are not published.","Hosted compute shares the same usage limits as model calls, so heavy machine use reduces model headroom.","Cloud environment availability depends on your organization's rollout and settings.","Usage limits are described only as multiples of Pro, not in fixed numbers."],"sources":[{"url":"https://factory.com/pricing","quote":"Bring your own inference","checked":"2026-11-14"},{"url":"https://factory.com/pricing","quote":"$100/mo per individual","checked":"2026-11-14"},{"url":"https://factory.com/pricing","quote":"$200/mo per individual","checked":"2026-11-14"},{"url":"https://factory.com/pricing","quote":"~10x the usage of Pro","checked":"2026-11-14"},{"url":"https://factory.com/pricing","quote":"per team, plus $40/mo per individual","checked":"2026-11-14"},{"url":"https://factory.com/pricing","quote":"All products; IDE, Web, CLI, Slack etc are included in all plans.","checked":"2026-11-14"},{"url":"https://docs.factory.com/pricing/individuals","quote":"sessions consume Standard Credits based on model usage, along with compute usage for cloud environments.","checked":"2026-11-14"},{"url":"https://docs.factory.com/pricing/individuals","quote":"Extra Usage: prepaid credits, $10 minimum, no expiration.","checked":"2026-11-14"},{"url":"https://docs.factory.com/droid-computers/overview","quote":"Platform-managed Droid Computers auto-pause when idle and auto-resume when a new session targets them.","checked":"2026-11-14"},{"url":"https://docs.factory.com/droid-computers/overview","quote":"They retain state: installed packages, files, running services, and configuration all persist between sessions.","checked":"2026-11-14"},{"url":"https://docs.factory.com/environments/overview","quote":"Factory builds a reusable snapshot with the code cloned and tools installed, so each new instance starts from the same prepared baseline.","checked":"2026-11-14"},{"url":"https://docs.factory.com/environments/overview","quote":"Availability depends on your organization's rollout and compute settings.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/factory","score":{"coding-agent":{"total":42,"rank":8,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"firecracker","name":"Firecracker","company":"Amazon Web Services","website":"https://firecracker-microvm.github.io","docsUrl":"https://github.com/firecracker-microvm/firecracker/tree/main/docs","pricingUrl":"https://github.com/firecracker-microvm/firecracker","categories":["code-sandbox"],"tagline":"Building block: the open-source KVM microVM monitor from AWS that many sandbox platforms run on.","summary":"Firecracker is a building block, not a ready-made sandbox service: a minimal virtual machine monitor that uses Linux KVM to start lightweight microVMs, built at AWS for Lambda and Fargate. Each microVM is configured through a REST API on the host with its vCPUs, memory, network interfaces, disks and rate limiters. Kata Containers and many agent sandbox platforms build on it. It is free Apache-2.0 software.","isolation":"KVM-based microVM per workload (Firecracker VMM)","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/firecracker-microvm/firecracker","lastRelease":"2026-11-09","sdks":["REST API (OpenAPI)"],"regions":"Runs on your own Linux hosts with KVM (bare metal or nested virtualization).","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own hardware","billedWhile":"Not applicable: no usage charges","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license.","notes":["A low-level virtual machine monitor: images, scheduling, networking setup and a user-facing API are left to whatever you build on top.","Default microVM size is 1 vCPU and 128 MiB of memory; both are configurable.","Host security depends on a correctly configured Linux host; the project publishes a production host setup guide."]},"bestFor":["Building your own microVM sandbox or serverless platform","Workloads that need VM isolation with fast startup and small memory overhead","Rate-limiting network and disk bandwidth per VM"],"watchOut":["Building block only: no SDK, image tooling or orchestration included","Needs Linux hosts with KVM","Secure multi-tenant use depends on careful host configuration"],"sources":[{"url":"https://github.com/firecracker-microvm/firecracker","quote":"Firecracker runs workloads in lightweight virtual machines, called microVMs","checked":"2026-11-13"},{"url":"https://github.com/firecracker-microvm/firecracker","quote":"The main component of Firecracker is a virtual machine monitor (VMM) that uses the Linux Kernel Virtual Machine (KVM) to create and run microVMs.","checked":"2026-11-13"},{"url":"https://github.com/firecracker-microvm/firecracker","quote":"Configure rate limiters for virtio devices which can limit the bandwidth, operations per second, or both.","checked":"2026-11-13"},{"url":"https://github.com/firecracker-microvm/firecracker","quote":"Firecracker is open sourced under Apache version 2.0.","checked":"2026-11-13"},{"url":"https://github.com/firecracker-microvm/firecracker/releases/tag/v1.18.0","quote":"v1.18.0","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/firecracker","score":{"code-sandbox":{"total":55,"rank":51,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"fly-sprites","name":"Fly.io Sprites","company":"Fly.io","website":"https://sprites.dev","docsUrl":"https://docs.fly.io/sprites/index.md","pricingUrl":"https://sprites.dev","categories":["dev-box","code-sandbox"],"tagline":"Persistent Linux microVMs that sleep when idle, billed on CPU time and memory actually used.","summary":"Sprites are hardware-isolated Linux VMs from Fly.io with a persistent 100 GB ext4 disk, checkpoints, and their own HTTPS URL. They pause automatically about 30 seconds after activity stops and wake on the next request. Billing is metered on cumulative CPU time ($0.0385/CPU-hour) and memory actually in use ($0.021875/GB-hour) only while a Sprite is running, plus hot storage while awake and cold storage for as long as data is kept. Optional monthly plans bundle CPU-hours, RAM GB-hours and storage and raise concurrency limits.","isolation":"Hardware-isolated microVM per Sprite","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","Go","Python","Elixir","CLI","REST API","Remote MCP server"],"regions":"Runs on Fly.io infrastructure; the pages checked do not document choosing a region for a Sprite.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.0385,"gibHour":0.021875,"instanceHour":null,"instance":null,"storageGibMonth":0.02,"billingIncrement":"metered usage (CPU time from cpu.stat, memory in use), rates quoted per hour","billedWhile":"Only while the Sprite is running (active); CPU on actual CPU time used and memory on actual memory used. Warm and cold (paused) Sprites pay only storage.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pay-as-you-go","monthlyFee":0,"includedUsageUsd":0,"limits":"Usage at standard rates; 10 Sprite creations per minute; community support. Concurrency limit for this tier not stated on the pricing page.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Adventurer","monthlyFee":20,"includedUsageUsd":0,"limits":"Includes 450 CPU-hours, 1,800 RAM GB-hours, 50 GB-months storage; 20 concurrent active and 20 warm Sprites; 60 creations/minute; community support.","includedVcpuHours":450,"includedGibHours":1800,"capped":false},{"name":"Veteran","monthlyFee":50,"includedUsageUsd":0,"limits":"Includes 800 CPU-hours, 3,200 RAM GB-hours, 100 GB-months storage; 50 concurrent active and 50 warm Sprites; community support.","includedVcpuHours":800,"includedGibHours":3200,"capped":false},{"name":"Hero","monthlyFee":100,"includedUsageUsd":0,"limits":"Includes 1,200 CPU-hours, 4,800 RAM GB-hours, 150 GB storage; 100 concurrently running and 100 warm Sprites, cold unlimited; standard email support.","includedVcpuHours":1200,"includedGibHours":4800,"capped":false},{"name":"Champion","monthlyFee":200,"includedUsageUsd":0,"limits":"Includes 1,800 CPU-hours, 7,200 RAM GB-hours, 225 GB-months storage; 200 concurrent active and 200 warm Sprites; standard email support.","includedVcpuHours":1800,"includedGibHours":7200,"capped":false},{"name":"Legend","monthlyFee":500,"includedUsageUsd":0,"limits":"Includes 3,200 CPU-hours, 12,500 RAM GB-hours, 400 GB-months storage; 500 concurrent active and 500 warm Sprites; standard email support.","includedVcpuHours":3200,"includedGibHours":12500,"capped":false},{"name":"Epic","monthlyFee":1000,"includedUsageUsd":0,"limits":"Includes 4,800 CPU-hours, 18,750 RAM GB-hours, 600 GB-months storage; 1,000 concurrent active and 1,000 warm Sprites; premium email support.","includedVcpuHours":4800,"includedGibHours":18750,"capped":false},{"name":"Mythic","monthlyFee":2000,"includedUsageUsd":0,"limits":"Includes 7,200 CPU-hours, 28,000 RAM GB-hours, 900 GB-months storage; 2,000 concurrent active and 2,000 warm Sprites; 240 creations/minute; premium email support; arranged via billing@fly.io.","includedVcpuHours":7200,"includedGibHours":28000,"capped":false}],"freeTierKind":"credit","freeTier":"Every new Fly.io account gets a one-time $1 trial credit (no credit card, lasts 7 days); invoices under $5 are waived.","notes":["CPU is billed on cumulative CPU time measured by cpu.stat, not on allocated vCPUs, and memory on actual usage, so an idle-but-awake Sprite costs little. Compute stops when the Sprite pauses (about 30 seconds after activity stops).","Storage: hot storage $0.000683/GB-hour (~$0.50/GB-month) bills only while awake; cold storage $0.000027/GB-hour (~$0.02/GB-month) bills for as long as the data exists. The storage rate we show is the $0.02 cold rate; add the hot rate for awake hours. Storage is billed on bytes written, not the 100 GB volume size.","Sprites bandwidth/egress is not metered.","Plan allowances are CPU-hours, RAM GB-hours and storage, not a dollar credit; our calculator subtracts them from usage. At list rates Hero's allowance is worth about $151 of compute (1,200 x $0.0385 + 4,800 x $0.021875). Usage above the allowance bills at standard rates on top of the fee. Fly.io account credits do not pay the plan fee.","Plan table ambiguity: the current sprites.dev FAQ confirms only Hero ($100/mo, 1,200 CPU-h, 4,800 RAM GB-h, 150 GB, 100 running/100 warm) and Mythic ($2,000/mo, 28,000 RAM GB-h) and names the other tiers without prices. The other fees and allowances come from Fly.io's 2026-01-16 changelog. Fly.io's 2026-01-22 release notes list only four tiers and price Legend at $200/mo instead of $500.","Billing increment is unclear: the pricing page says resources are 'billed hourly, based on actual usage', while the docs overview says 'Per-second' billing.","Limits: 8 vCPUs and a 100 GB disk per Sprite (fixed). Memory is managed and autoscaled by the platform with no published fixed figure; Sprites release notes say memory starts at 2 GB and can autoscale up to 16 GB. We show max memory as not stated.","Concurrency limits are per org and depend on the plan; creation rate is 10/minute on pay-as-you-go. A Sprite stays awake only while there is activity; the Tasks API can hold it awake (max 1 hour per task, renewable)."]},"bestFor":["Agents and dev environments that sit idle most of the time, since paused Sprites pay only storage","Workloads with low average CPU use, because CPU is billed on CPU time actually consumed","Persistent per-user or per-agent Linux boxes with checkpoints and their own HTTPS URL"],"watchOut":["Fixed 8 vCPU and 100 GB disk per Sprite; memory has no published fixed size","No GPU support is described","Code running inside a Sprite does not keep it awake by itself; it pauses after about 30 seconds without requests, output, open TCP connections or an active task","Plan allowances are unit bundles, and plan fees are not covered by account credits"],"sources":[{"url":"https://sprites.dev","quote":"CPU time | Cumulative CPU usage measured by cpu.stat | $0.0385 / CPU-hour","checked":"2026-11-14"},{"url":"https://sprites.dev","quote":"Memory time | Actual memory usage | $0.021875 / GB-hour","checked":"2026-11-14"},{"url":"https://sprites.dev","quote":"Hero ($100/mo) includes 1,200 CPU-hours, 4,800 RAM GB-hours, and 150 GB of storage","checked":"2026-11-14"},{"url":"https://sprites.dev","quote":"Every new Fly.io account gets a $1 trial credit, no credit card required. It works across Sprites, Machines and Managed Postgres, and lasts 7 days.","checked":"2026-11-14"},{"url":"https://sprites.dev","quote":"No. Sprites bandwidth isn't metered today.","checked":"2026-11-14"},{"url":"https://fly.io/pricing/","quote":"Cold | Bills for as long as you keep it | $0.000027 / GB-hour | $0.02","checked":"2026-11-14"},{"url":"https://fly.io/changelog/new-sprites-plans/","quote":"Adventurer Level 20 | $20.00 | 20 | 20 | 450 | 1800 | 50 | Community","checked":"2026-11-14"},{"url":"https://fly.io/sprites/release-notes/","quote":"Level 200 (Legend) - $200/mo","checked":"2026-11-14"},{"url":"https://docs.fly.io/sprites/concepts/lifecycle.md","quote":"Each Sprite runs with: 8 vCPUs ... 100 GB of storage","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/fly-sprites","score":{"dev-box":{"total":70,"rank":10,"of":38},"code-sandbox":{"total":69,"rank":18,"of":63}},"comparablePrice":{"dev-box":20,"code-sandbox":19.4656},"benchmarks":[]},{"slug":"freestyle","name":"Freestyle","company":"Freestyle","website":"https://www.freestyle.sh","docsUrl":"https://www.freestyle.sh/docs/vms","pricingUrl":"https://www.freestyle.sh/pricing","categories":["dev-box","code-sandbox"],"tagline":"Full Linux VMs for agent and app-builder workloads, with pause/resume, snapshots, forking and private networks.","summary":"Freestyle sells full Linux VMs aimed at long-running agent tasks and AI app builders, with commands, files, PTY sessions, domains, firewalls and VPCs managed through a TypeScript SDK and CLI. VMs can be paused with memory intact, snapshotted and started from snapshots, and resized up while running. vCPU, memory and storage are metered per second on the amount allocated, and every plan includes the same monthly compute allowance. Hobby and Pro have monthly minimums that count as usage credit.","isolation":"Hardware-virtualized VM (supports nested virtualization; hypervisor not named)","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","CLI"],"regions":"Not published","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.04032,"gibHour":0.0129,"instanceHour":null,"instance":null,"storageGibMonth":0.06278,"billingIncrement":"per second (rates published per hour)","billedWhile":"wall-clock on allocated vCPU and memory while running, whether or not the guest is busy; storage is billed while the VM exists, including when paused","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Monthly allowance of 200 vCPU-hours, 400 GiB-hours memory, 60,000 GiB-hours storage, 50 GB transfer, as hard stops (VMs pause until reset); 10 concurrent VMs, 10 saved VMs, 10 snapshots; max 4 vCPU / 8 GiB / 32 GB per VM","includedVcpuHours":200,"includedGibHours":400,"capped":true},{"name":"Hobby","monthlyFee":50,"includedUsageUsd":50,"limits":"Same monthly compute allowance as Free plus $50 of usage covered by the fee; 500 GB transfer; 40 concurrent VMs, 200 saved VMs, 1,000 snapshots; max 8 vCPU / 16 GiB / 64 GB per VM","includedVcpuHours":200,"includedGibHours":400,"capped":false},{"name":"Pro","monthlyFee":500,"includedUsageUsd":500,"limits":"Same monthly compute allowance as Free plus $500 of usage covered by the fee; 500 GB transfer; 400 concurrent VMs, 4,000 saved VMs, 12,000 snapshots; max 32 vCPU / 64 GiB / 256 GB per VM","includedVcpuHours":200,"includedGibHours":400,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan with a monthly allowance (same on every plan) of 200 vCPU-hours, 400 GiB-hours of memory, 60,000 GiB-hours of storage, plus 50 GB data transfer; no usage charges, allowances are hard stops. Adding a card charges a refunded $1 verification.","notes":["Storage is published as $0.000086 per GiB-hour; 0.000086 x 730 hours = $0.06278 per GiB-month.","Compute is billed on allocated vCPU and memory, not on measured CPU use. Active-CPU pricing is offered only on Enterprise.","Monthly allowances apply on every plan before any charge: 200 vCPU-hours ($8.064), 400 GiB-hours of memory ($5.16) and 60,000 GiB-hours of storage ($5.16), about $18.38 a month at list price. Our calculator subtracts these allowances before billing.","On Hobby and Pro you pay the greater of the plan fee or usage: the $50 / $500 fee covers usage dollar for dollar, then overage is billed at the listed rates.","Data transfer is $0.02 per GB after 50 GB (Free) or 500 GB (Hobby/Pro), counted in both directions across the datacenter boundary; traffic between your VMs inside a VPC is free.","A paused VM releases its compute and bills only for storage; it does not count toward concurrent VMs but does count as a saved VM. An idle timeout can pause VMs with no network activity.","Free plan maximum per VM is the default size (4 vCPU, 8 GiB, 32 GB). The pricing page marks Persistent VMs, Persistent Snapshots and Custom Sizing as unavailable on Free and lists disk in GiB, while the Pricing and Limits doc marks Persistent VMs available on every plan, allows smaller sizes on Free and lists disk in GB.","GPU sandboxes are offered privately to enterprise customers only, with no published prices; gpu is set to false for self-serve plans.","The pricing page lists only VM usage; no separate prices are published for other Freestyle products."]},"bestFor":["Long-running agent workspaces that pause with memory intact and resume later","AI app builders that need per-user VMs with domains, firewalls and private VPC networking","Branching or cloning a VM state via cheap snapshots"],"watchOut":["Allocated vCPU and memory bill while running even when the guest is idle; set an idle timeout to pause","Host resources are oversubscribed; the vendor advises against sustained CPU-heavy work such as large builds","vm.exec() has a 5-minute maximum timeout; longer jobs need a PTY session","Resize is grow-only; to shrink a VM you must create a new one"],"sources":[{"url":"https://www.freestyle.sh/pricing","quote":"Included amounts reset monthly. The compute allowances are the same on every plan — Free, Hobby, and Pro.","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/pricing","quote":"The default VM size is 4 vCPUs and 8 GiB of memory, with up to 32 GiB of disk.","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/vms/pricing-and-limits","quote":"| Hour of vCPU | $0.04032 | 200 per month |","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/vms/pricing-and-limits","quote":"A paused VM holds no compute reservation, so it bills only for storage while it sleeps.","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/vms/pricing-and-limits","quote":"| **Hobby** | $50/month minimum, including $50 usage | Up to 40 |","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/vms/lifecycle","quote":"Resizing is up-only: none of `cpu`, `memory`, or `storage` can be reduced.","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/vms","quote":"The maximum timeout is five minutes; use a PTY session for longer interactive work.","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/onboard.md","quote":"Freestyle offers private GPU sandboxes for enterprise customers","checked":"2026-11-14"},{"url":"https://www.freestyle.sh/docs/onboard.md","quote":"We also oversubscribe host resources.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/freestyle","score":{"dev-box":{"total":63,"rank":18,"of":38},"code-sandbox":{"total":61,"rank":39,"of":63}},"comparablePrice":{"dev-box":50,"code-sandbox":50},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":1548,"unit":"ms","better":"lower","date":"2026-11-13","rank":20,"of":27,"note":"99% of starts succeeded"},{"source":"aisandboxbench","metric":"Cold start (median)","value":0.693,"unit":"s","better":"lower","date":"2026-10-23","rank":3,"of":9},{"source":"aisandboxbench","metric":"Agent tool-call replay (median)","value":24.072,"unit":"s","better":"lower","date":"2026-10-23","rank":6,"of":9}]},{"slug":"github-actions-runners","name":"GitHub-hosted runners","company":"GitHub","website":"https://github.com/features/actions","docsUrl":"https://docs.github.com/en/actions/concepts/runners/github-hosted-runners","pricingUrl":"https://docs.github.com/en/billing/reference/actions-runner-pricing","categories":["ci-runner"],"tagline":"GitHub's own Actions runners: a fresh VM per job, $0.006 a minute for 2-core Linux, free on public repositories.","summary":"GitHub-hosted runners run each GitHub Actions job on a new virtual machine (single-CPU runners use a container on a shared VM), with Linux, Windows and macOS images maintained by GitHub. In private repositories usage is priced per minute, rounded up to the whole minute, and drawn first from a monthly dollar allowance included with every GitHub plan. Standard runners are free in public repositories. Larger runners up to 96 cores, GPU runners and custom images are available on Team and Enterprise Cloud.","isolation":"New virtual machine per job (single-CPU runners share a VM as containers)","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["GitHub Actions workflows","REST API","GitHub CLI"],"regions":"Linux jobs run on a multi-cloud pool across AWS and Microsoft Azure; Windows and macOS runners run in Azure. Region choice for private networking is available on larger runners.","features":{"pauseResume":false,"snapshots":true,"gpu":true,"customImages":true,"networkControls":null,"maxSessionHours":6,"maxVcpu":96,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.36,"instance":"Standard Linux 2-core x64 runner for private repositories: 2 vCPU, 8 GB ($0.006 per minute)","storageGibMonth":0.07,"billingIncrement":"per minute; each job is rounded up to the nearest whole minute","billedWhile":"Only while a workflow job runs on the runner; standard runners are free in public repositories.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[{"type":"Linux 4-core GPU runner (GPU model not stated)","perHour":3.12}],"plans":[{"name":"GitHub Free","monthlyFee":0,"includedUsageUsd":12,"limits":"$12 of included compute a month (2,000 minutes on standard Linux), 500 MB artifact storage, 10 GB cache per repository. Without a payment method, jobs stop once the allowance is used. Larger runners are not available.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"GitHub Team","monthlyFee":4,"includedUsageUsd":18,"limits":"$4 per user a month; $18 of included compute a month (3,000 minutes on standard Linux), 2 GB artifact storage, 75 GB custom image storage. Larger runners available and can draw from the allowance.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"GitHub Enterprise Cloud","monthlyFee":21,"includedUsageUsd":300,"limits":"$21 per user a month; $300 of included compute a month (50,000 minutes on standard Linux), 50 GB artifact storage, 150 GB custom image storage.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Standard runners are free and unlimited in public repositories. For private repositories every plan includes a monthly compute allowance: $12 on GitHub Free (2,000 standard Linux minutes).","notes":["Rate shown is the standard Linux 2-core x64 runner for private repositories (2 vCPU, 8 GB): $0.006 per minute x 60 = $0.36 per hour. In public repositories ubuntu-latest gets 4 vCPU / 16 GB at no charge.","Other rates per minute: Linux 1-core $0.002, Linux 2-core arm64 $0.005, Windows 2-core $0.010, macOS 3- or 4-core $0.062. Larger Linux x64 runners run from 4-core ($0.012) to 96-core ($0.252); they are not linear in size.","GPU rate: Linux 4-core GPU runner $0.052 per minute x 60 = $3.12 per hour (Windows 4-core GPU $0.102 per minute). The pricing page does not name the GPU model.","The included allowance is a dollar amount shared with cloud sandbox compute and Copilot code review runner usage; plan fees are per user per month. GitHub Pro for individuals includes $18 of compute.","Storage rate shown is Actions cache beyond the free 10 GB per repository ($0.07 per GB-month); artifact and Packages storage is $0.25 per GB-month and custom image storage $0.07 per GB-month.","Each job can run for at most 6 hours. Larger runners, custom images and GPU runners need GitHub Team or Enterprise Cloud and are billed even in public repositories.","Custom images are created with a snapshot step in a workflow job and are billed as storage."]},"bestFor":["Public open-source repositories, where standard runners are free","Small private projects that fit inside the monthly included allowance","Teams that want Linux, Windows, macOS and GPU runners without another vendor"],"watchOut":["Per-minute rates are higher than most third-party runner services","Every job is rounded up to a whole minute","Jobs are capped at 6 hours","Larger and GPU runners need a paid organization plan and are billed in public repositories too"],"sources":[{"url":"https://docs.github.com/en/billing/reference/actions-runner-pricing","quote":"Linux 2-core (x64) | actions_linux | $0.006","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/reference/actions-runner-pricing","quote":"GitHub rounds the minutes and partial minutes each job uses up to the nearest whole minute.","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/reference/actions-runner-pricing","quote":"Linux 4-core | linux_4_core_gpu | $0.052","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-actions","quote":"GitHub Free | 500 MB | $12 USD | 10 GB | Not applicable","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-actions","quote":"As a guide, $300 USD covers 50,000 minutes on a standard Linux runner at $0.006 USD per minute.","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-actions","quote":"Actions cache | $0.07 USD","checked":"2026-11-13"},{"url":"https://github.com/pricing","quote":"$ 4 USD per user / month","checked":"2026-11-13"},{"url":"https://github.com/pricing","quote":"$ 21 USD per user / month","checked":"2026-11-13"},{"url":"https://docs.github.com/en/actions/concepts/runners/github-hosted-runners","quote":"With the exception of single-CPU runners, each GitHub-hosted runner is a new virtual machine (VM) hosted by GitHub.","checked":"2026-11-13"},{"url":"https://docs.github.com/en/actions/concepts/runners/github-hosted-runners","quote":"These jobs can run on either Amazon Web Services (AWS) or Microsoft Azure.","checked":"2026-11-13"},{"url":"https://docs.github.com/en/actions/reference/limits","quote":"Each job in a workflow can run for up to 6 hours of execution time.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/github-actions-runners","score":{"ci-runner":{"total":70,"rank":5,"of":10}},"comparablePrice":{"ci-runner":21},"benchmarks":[]},{"slug":"github-codespaces","name":"GitHub Codespaces","company":"GitHub","website":"https://github.com/features/codespaces","docsUrl":"https://docs.github.com/en/codespaces","pricingUrl":"https://docs.github.com/en/billing/concepts/product-billing/github-codespaces","categories":["dev-box"],"tagline":"Cloud dev environments on a dedicated VM per codespace, billed per core-hour, with a free monthly quota for individuals.","summary":"GitHub Codespaces creates a development environment for a repository on a dedicated virtual machine, running a dev container that you open in VS Code, the browser or over the GitHub CLI. Compute is billed per hour while a codespace is active, in proportion to its cores: $0.18 an hour for 2 cores up to $2.88 for 32 cores, plus $0.07 per GB-month of storage while it exists. Personal GitHub Free accounts get 120 core-hours and 15 GB-months each month. Codespaces stop after an idle timeout and have a maximum lifetime of 12 hours before they must be restarted.","isolation":"Dedicated, private VM per codespace running a dev container","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["GitHub CLI","REST API","VS Code"],"regions":"Set automatically from your location, or chosen manually in settings; the region list is not on the pages checked.","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":true,"networkControls":null,"maxSessionHours":12,"maxVcpu":32,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.18,"instance":"2-core machine","storageGibMonth":0.07,"billingIncrement":"Compute time reported hourly and billed monthly","billedWhile":"Compute while the codespace is active; storage for as long as the codespace or prebuild exists.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"GitHub Free (personal account)","monthlyFee":0,"includedUsageUsd":0,"limits":"120 core-hours of compute and 15 GB-months of storage each month. Usage beyond it is billed if a payment method is on file, otherwise blocked.","includedVcpuHours":120,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Personal accounts: 120 core-hours and 15 GB-months a month on GitHub Free, 180 core-hours and 20 GB-months on GitHub Pro. Organization plans include no free quota.","notes":["Compute is linear in cores: 2-core $0.18, 4-core $0.36, 8-core $0.72, 16-core $1.44, 32-core $2.88 per hour. The included quota is in core-hours, so 120 core-hours is 60 hours on a 2-core machine.","GitHub prices by core and does not list RAM per machine type on the billing page, so we leave memory unset.","Prebuilds use GitHub Actions compute and their storage counts toward codespace storage.","A codespace stops after 30 minutes idle by default (settable from 5 to 240 minutes) and has a 12-hour maximum lifetime even while in use. Stopping keeps saved files but ends running processes.","GitHub Pro's price is not on the pricing pages checked, so it is not listed as a plan here."]},"bestFor":["Repository-based dev environments defined with a dev container","Personal projects within the monthly free core-hours","Teams already billing through GitHub"],"watchOut":["Codespaces stop after at most 12 hours and need restarting","Organization accounts get no free quota","Per-core pricing is higher than most agent-focused dev box providers","Not designed as an API sandbox for agents"],"sources":[{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-codespaces","quote":"| Codespaces compute | 2 core       | 1 hour          | 2                         | $0.18 |","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-codespaces","quote":"| Codespaces storage | Storage      | 1 GB-month      | Not applicable            | $0.07 |","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-codespaces","quote":"| GitHub Free for personal accounts | 15 GB-month       | 120 hrs                |","checked":"2026-11-13"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/github-codespaces","quote":"The compute cost is proportional to the number of processor cores in the machine type you choose for your codespace","checked":"2026-11-13"},{"url":"https://docs.github.com/en/get-started/learning-about-github/githubs-plans","quote":"120 GitHub Codespaces core hours per month","checked":"2026-11-13"},{"url":"https://docs.github.com/en/codespaces/about-codespaces/understanding-the-codespace-lifecycle","quote":"A codespace has a maximum lifetime of 12 hours, regardless of your idle timeout policy or settings.","checked":"2026-11-13"},{"url":"https://docs.github.com/en/codespaces/about-codespaces/deep-dive","quote":"The VM is both dedicated and private to you.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/github-codespaces","score":{"dev-box":{"total":63,"rank":19,"of":38}},"comparablePrice":{"dev-box":39.6},"benchmarks":[]},{"slug":"github-copilot-coding-agent","name":"GitHub Copilot coding agent","company":"GitHub","website":"https://github.com/features/copilot","docsUrl":"https://docs.github.com/en/copilot/concepts/security-governance-and-network-settings/about-cloud-sandboxes","pricingUrl":"https://github.com/features/copilot/plans","categories":["coding-agent"],"tagline":"Copilot cloud sessions in GitHub-hosted sandboxes, billed per vCPU-second and GiB-second on top of the Copilot plan.","summary":"Copilot in the cloud (the Copilot coding agent) runs sessions in an isolated GitHub-hosted Linux sandbox by default, or in a GitHub Actions job when a repository needs custom setup steps, larger or self-hosted runners, or Windows. Unlike most coding agents, the machine is billed separately from AI usage: sandbox compute costs $0.000024 per vCPU-second and memory $0.000003 per GiB-second while running, with snapshot storage for stopped sessions. That compute draws first on the monthly included compute allowance of your GitHub account ($12 on GitHub Free). Cloud sessions need a paid Copilot plan, from $10 a month.","isolation":"Isolated GitHub-hosted Linux environment per session; the isolation technology is not stated. Alternatively a GitHub Actions runner job.","isolationKind":"unknown","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Web app","GitHub Mobile","VS Code","Copilot CLI","Copilot SDK","Slack and Microsoft Teams"],"regions":"Not stated.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":false,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.0864,"gibHour":0.0108,"instanceHour":null,"instance":null,"storageGibMonth":0.005,"billingIncrement":"per second","billedWhile":"From session start until it is stopped or deleted: compute and allocated memory only while running. Stopped sessions pay snapshot storage until deleted.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Copilot Pro","monthlyFee":10,"includedUsageUsd":12,"limits":"Includes Copilot in the cloud and code review, with $15 of GitHub AI Credits a month for model usage ($10 base plus $5 flex). Sandbox compute is billed separately and draws first on the GitHub account's included compute usage: $12 a month on GitHub Free accounts (shown here), $18 on GitHub Pro.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Copilot Pro+","monthlyFee":39,"includedUsageUsd":12,"limits":"$70 of AI Credits a month, premium models including Opus, delegation to third-party agents. Sandbox compute draws on the GitHub account's $12 (Free) or $18 (Pro) monthly included compute usage.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Copilot Max","monthlyFee":100,"includedUsageUsd":12,"limits":"$200 of AI Credits a month and priority access to new models. Sandbox compute draws on the GitHub account's $12 (Free) or $18 (Pro) monthly included compute usage.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"None for cloud sessions: Copilot Free does not include Copilot in the cloud. Paid Copilot users get a monthly included compute amount from their GitHub account plan ($12 on GitHub Free, $18 on GitHub Pro) that covers sandbox compute and memory.","notes":["The cloud machine is priced separately from AI usage: compute $0.000024 per vCPU-second (= $0.0864 per vCPU-hour), memory $0.000003 per GiB-second (= $0.0108 per GiB-hour) on allocated memory, and snapshot storage $0.005 per GiB-month for stopped sessions.","Ambiguity: the cloud sandbox page lists the compute unit as 'vCPU second', while the billing page calls it 'Compute second' and describes it as the time a session is running. We treat it as per vCPU. The sandbox's vCPU count and memory size are not published.","Sandbox compute and memory share the GitHub account's included compute usage with Actions standard runners in private repositories; after it is used up, usage is billed if a payment method is on file, otherwise it is blocked. Budgets can stop usage at a limit. Cloud sandboxes are not part of AI credit budgets.","Sessions can instead run in a GitHub Actions job, billed as Actions runner usage. Actions-based sessions are limited to 59 minutes and support custom setup steps, larger and self-hosted runners, and Windows; cloud sandboxes do not support custom compute or setup steps. No run-time limit is stated for cloud sandbox sessions.","Stopped sessions resume later with files and in-progress work from a snapshot. Network access follows the repository and organization firewall allowlist.","Copilot Business is $19 and Enterprise $39 per granted seat per month with pooled AI credits; GitHub Team and Enterprise accounts have larger included compute amounts (Team $18)."]},"bestFor":["Teams already on GitHub that want agents working from issues and pull requests.","Short tasks where per-second machine billing and the free monthly compute amount keep machine cost low.","Repositories that need Windows or self-hosted runners, using the Actions-based option."],"watchOut":["The sandbox is billed on top of the Copilot plan and AI credits once the account's included compute is used.","Sandbox size is not published, and custom compute or setup steps need the Actions option.","Actions-based sessions stop after 59 minutes.","Not available on Copilot Free."],"sources":[{"url":"https://docs.github.com/en/copilot/concepts/security-governance-and-network-settings/about-cloud-sandboxes","quote":"| Compute | Time that a cloud sandbox session is running.                    | vCPU second | $0.000024   |","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/concepts/security-governance-and-network-settings/about-cloud-sandboxes","quote":"| Memory  | Memory allocated to a cloud sandbox session while it is running. | GiB second  | $0.000003   |","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/concepts/security-governance-and-network-settings/about-cloud-sandboxes","quote":"| Storage | Snapshot storage for stopped sessions.                           | GiB month   | $0.005      |","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/concepts/security-governance-and-network-settings/about-cloud-sandboxes","quote":"A cloud sandbox is an isolated Linux environment that GitHub creates for a Copilot session.","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/concepts/security-governance-and-network-settings/about-cloud-sandboxes","quote":"While a session is stopped, the cloud sandbox keeps a snapshot of its state so you can pick up where you left off.","checked":"2026-11-14"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/cloud-and-local-sandboxes","quote":"Compute | Time that a cloud sandbox session is running.                    | Compute second | $0.000024","checked":"2026-11-14"},{"url":"https://docs.github.com/en/billing/concepts/product-billing/cloud-and-local-sandboxes","quote":"Memory is measured based on the memory allocated to the session, not the memory actively in use.","checked":"2026-11-14"},{"url":"https://docs.github.com/en/billing/reference/product-usage-included","quote":"The included compute usage is expressed as a monetary amount, not a quantity of minutes or other usage units.","checked":"2026-11-14"},{"url":"https://docs.github.com/en/billing/reference/product-usage-included","quote":"Included compute usage (per month)","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/concepts/agents/coding-agent/about-coding-agent","quote":"Each Copilot session that runs on GitHub Actions has a maximum execution time of 59 minutes.","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/concepts/agents/coding-agent/about-coding-agent","quote":"In the Copilot in the cloud experience, cloud and GitHub Actions compute draw on the same monthly GitHub account allowance, separate from AI credits.","checked":"2026-11-14"},{"url":"https://github.com/features/copilot/plans","quote":"Access to Copilot in the cloud and code review","checked":"2026-11-14"},{"url":"https://github.com/features/copilot/plans","quote":"$15 monthly total credits for Pro","checked":"2026-11-14"},{"url":"https://github.com/features/copilot/plans","quote":"$200 monthly total credits for Max","checked":"2026-11-14"},{"url":"https://docs.github.com/en/copilot/get-started/plans","quote":"Copilot Free users can continue to use Copilot Chat on GitHub.com and GitHub Mobile, and can upgrade to a paid plan to access Copilot in the cloud.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/github-copilot-coding-agent","score":{"coding-agent":{"total":42,"rank":9,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"google-cloud-workstations","name":"Google Cloud Workstations","company":"Google Cloud","website":"https://cloud.google.com/workstations","docsUrl":"https://cloud.google.com/workstations/docs/overview","pricingUrl":"https://cloud.google.com/workstations/pricing","categories":["dev-box"],"tagline":"Managed dev environments on Compute Engine VMs in your own project, billed at VM rates plus a per-vCPU management fee.","summary":"Cloud Workstations runs each developer's environment as a container on a Compute Engine VM created in your own Google Cloud project and VPC, with a persistent disk kept as the home directory between sessions. You pay Compute Engine rates for the VM and disk, a management fee of $0.05 per vCPU-hour while a workstation runs, and $0.20 an hour per cluster control plane. An e2-standard-2 workstation (2 vCPU, 8 GiB) costs $0.16701142 an hour in us-central1 including the management fee. GPU machine types and custom container images are supported.","isolation":"Compute Engine VM per running workstation, with the workstation as a container on it","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["gcloud CLI","REST API","Console"],"regions":"Around 30 Google Cloud regions across the Americas, Europe, the Middle East and Asia-Pacific; prices vary by region.","features":{"pauseResume":true,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":96,"maxMemoryGib":768},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.16701142,"instance":"e2-standard-2: 2 vCPU, 8 GiB (us-central1, VM plus management fee)","storageGibMonth":0.1,"billingIncrement":"Compute Engine billing; management fee metered from start until stop","billedWhile":"VM and management fee while the workstation runs; persistent disk while it exists; the cluster control plane every hour whether workstations are in use or not.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"credit","freeTier":"New Google Cloud customers get $300 in free credit; there is no free allowance specific to Cloud Workstations.","notes":["Each workstation cluster (control plane) costs $0.20 an hour, about $144 for a 30-day month, regardless of use. Most setups need one cluster per region or VPC; it is not included in the hourly price we show.","E2 standard sizes are linear in 2 vCPU / 8 GiB units: e2-standard-4 $0.33402284, e2-standard-32 $2.67218272 an hour. The shared-core e2-medium (1 vCPU, 4 GiB) is $0.08350571. These totals include the $0.05 per vCPU-hour management fee.","Storage shown is balanced persistent disk at $0.1 per GB-month, the rate used in Google's own pricing example; other disk types cost differently.","GPU workstations are offered as A2 machine types, from a2-highgpu-1g (1 GPU, 12 vCPU, 85 GiB) at $4.273385 an hour total.","Largest listed types are 96 vCPU (n1-standard-96, n2d-highmem-96 with 768 GiB). Committed and sustained use discounts apply to the VM portion.","Workstations stop after a configurable idle timeout; the VM is deleted at the end of a session and the home disk is kept."]},"bestFor":["Companies that need dev environments inside their own Google Cloud project and VPC","Teams that want admin-controlled custom images and policies for every developer","GPU-backed development environments"],"watchOut":["The $0.20 an hour cluster fee makes it costly for one or two users","Setup requires a Google Cloud project, cluster and workstation configuration","Processes do not survive a stop: only the home disk persists","Prices vary by region and machine family"],"sources":[{"url":"https://cloud.google.com/workstations/pricing","quote":"|e2-standard-2 |2 |8 GiB |$0.16701142 / 1 hour |","checked":"2026-11-13"},{"url":"https://cloud.google.com/workstations/pricing","quote":"Workstations accrue a management fee of $0.05 per virtual CPU, per hour.","checked":"2026-11-13"},{"url":"https://cloud.google.com/workstations/pricing","quote":"Workstation clusters accrue a control plane fee of $0.20 per cluster, per hour, regardless of whether the cluster's workstations are in use.","checked":"2026-11-13"},{"url":"https://cloud.google.com/workstations/pricing","quote":"In addition, add the price of Persistent Disk resources: 100GB * $0.1 per GB = $10.00 per month","checked":"2026-11-13"},{"url":"https://cloud.google.com/workstations/pricing","quote":"|n2d-highmem-96 |96 |768 GiB |$10.27104 / 1 hour |","checked":"2026-11-13"},{"url":"https://cloud.google.com/workstations/pricing","quote":"|a2-highgpu-1g |1 |12 |85 GiB |$4.273385 / 1 hour |","checked":"2026-11-13"},{"url":"https://cloud.google.com/workstations/docs/architecture","quote":"a Compute Engine VM is dynamically created inside your project and VPC after a workstation is started.","checked":"2026-11-13"},{"url":"https://cloud.google.com/free","quote":"New customers get $300 in free credit to try Google Cloud products and build a proof of concept.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/google-cloud-workstations","score":{"dev-box":{"total":66,"rank":15,"of":38}},"comparablePrice":{"dev-box":36.742512399999995},"benchmarks":[]},{"slug":"gvisor","name":"gVisor","company":"Google","website":"https://gvisor.dev","docsUrl":"https://gvisor.dev/docs/","pricingUrl":"https://github.com/google/gvisor","categories":["code-sandbox"],"tagline":"Building block: Google's open-source application kernel that sandboxes containers through an OCI runtime, runsc.","summary":"gVisor is a building block, not a ready-made sandbox service: an application kernel written in Go that runs in userspace and implements a Linux-like interface, so sandboxed containers do not talk to the host kernel directly. Its OCI runtime, runsc, plugs into Docker, containerd and Kubernetes. Many sandbox products and GKE Sandbox use it underneath. It is free Apache-2.0 software for x86_64 and ARM64 Linux.","isolation":"Userspace application kernel (gVisor) between the container and the host kernel","isolationKind":"gvisor","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/google/gvisor","lastRelease":"2026-11-11","sdks":["OCI runtime (runsc)","Docker","containerd","Kubernetes RuntimeClass"],"regions":"Runs on your own x86_64 or ARM64 Linux hosts.","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own hardware","billedWhile":"Not applicable: no usage charges","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license.","notes":["A low-level isolation layer: you still need your own orchestration, API, networking and lifecycle management to build a sandbox service on it.","Runs standard OCI container images through Docker or Kubernetes."]},"bestFor":["Hardening containers that run untrusted code without moving to full VMs","Adding a sandboxed runtime class to an existing Docker or Kubernetes setup","Teams building their own sandbox platform"],"watchOut":["Building block only: no API, SDK or sandbox lifecycle features on its own","Not a VM; it reduces but does not remove the host kernel attack surface","Linux only, x86_64 and ARM64"],"sources":[{"url":"https://github.com/google/gvisor","quote":"It is an application kernel that implements a Linux-like interface.","checked":"2026-11-13"},{"url":"https://github.com/google/gvisor","quote":"The runsc runtime integrates with Docker and Kubernetes, making it simple to run sandboxed containers.","checked":"2026-11-13"},{"url":"https://github.com/google/gvisor","quote":"gVisor builds on x86_64 and ARM64.","checked":"2026-11-13"},{"url":"https://github.com/google/gvisor/releases/tag/release-20261109.0","quote":"release-20261109.0","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/gvisor","score":{"code-sandbox":{"total":54,"rank":53,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"hopx","name":"Hopx","company":"Bunnyshell","website":"https://hopx.ai","docsUrl":"https://docs.hopx.ai/","pricingUrl":"https://hopx.ai/#pricing","categories":["code-sandbox","dev-box"],"tagline":"Firecracker microVM sandboxes from Bunnyshell, billed per second for vCPU, memory and storage.","summary":"Hopx runs each sandbox in a Firecracker microVM created from a template, driven from Python or JavaScript SDKs, a CLI, a REST API or an MCP server. Sandboxes have no execution time limit unless you set a timeout, and can be paused with memory kept or stopped with the filesystem kept. Usage is billed per second: $0.000014 per vCPU-second and $0.0000045 per GiB-second of memory, plus storage. New accounts get $200 in free credits with no credit card.","isolation":"Firecracker microVM per sandbox","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","CLI","REST API","MCP server"],"regions":"Region chosen per sandbox; the docs show us-east and eu-west as examples and do not publish a full list.","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":0.07884,"billingIncrement":"per second","billedWhile":"While the sandbox is running ('pay only when running'). The pricing section does not say whether paused or stopped sandboxes pay for storage.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"credit","freeTier":"$200 in free credits for new accounts, no credit card required. The site does not say whether the credits expire.","notes":["Rates are published per second: vCPU $0.000014/s ($0.0504 per vCPU-hour), memory $0.0000045 per GiB-second ($0.0162 per GiB-hour), storage $0.00000003 per GiB-second ($0.000108 per GiB-hour).","Storage is published per second only; the monthly figure shown assumes a 730-hour month ($0.07884 per GiB-month). With a 30-day month it would be $0.07776.","No plan table is published on the website. The CLI reference shows a sample 'hopx usage plans' output listing Free ($0), Pro ($29/mo) and Enterprise (Custom), but it is example output and the website does not confirm these plans or prices.","Resource limits per sandbox are not published as fixed numbers. The template docs say limits depend on your plan and give typical ranges of 1-64 CPU cores, 512 MB-64 GB memory and 1-250 GB disk.","Pausing keeps the full state including memory; stopping keeps the filesystem but loses runtime state. Billing for paused or stopped sandboxes is not described.","Network control: a sandbox can be created with internet access turned off. Finer egress rules are not described in the pages checked.","GPU sandboxes are not mentioned on the website or in the docs index."]},"bestFor":["Running untrusted or model-generated code in microVMs from Python or JavaScript","Long-running agent or batch jobs, since sandboxes have no execution time limit by default","Desktop automation, with VNC, screenshots, mouse and keyboard control in the SDK"],"watchOut":["No published plans, concurrency limits or per-sandbox size limits; limits are said to depend on the account plan","Storage price is only given per GiB-second, and billing for paused or stopped sandboxes is not stated","Region list is not published; the docs only show us-east and eu-west as examples"],"sources":[{"url":"https://hopx.ai","quote":"vCPU $0.00001400/s","checked":"2026-11-14"},{"url":"https://hopx.ai","quote":"GiB $0.00000450/s","checked":"2026-11-14"},{"url":"https://hopx.ai","quote":"GiB $0.00000003/s","checked":"2026-11-14"},{"url":"https://hopx.ai","quote":"Get started with $200 in free credits.","checked":"2026-11-14"},{"url":"https://hopx.ai","quote":"No credit card required • Free $200 credits • Cancel anytime","checked":"2026-11-14"},{"url":"https://hopx.ai","quote":"No execution time limits. Keep agents, notebooks, or jobs running for hours, days, or weeks — with full state persistence.","checked":"2026-11-14"},{"url":"https://hopx.ai","quote":"Firecracker microVMs provide hardware-level security and kernel isolation","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/core-concepts/sandboxes/managing-state","quote":"When you pause a sandbox, its state is preserved. When you resume it, everything continues from where it left off. Stopped sandboxes lose their runtime state but can be restarted.","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/core-concepts/sandboxes/timeout","quote":"timeout_seconds not specified = no timeout","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/core-concepts/sandboxes/creating","quote":"Control whether the sandbox has internet access:","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/core-concepts/sandboxes/creating","quote":"region=\"eu-west\"  # or \"us-east\", etc.","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/core-concepts/templates/configuration","quote":"# Resource limits depend on your plan","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/core-concepts/templates/building","quote":"Build custom sandbox templates from Docker images","checked":"2026-11-14"},{"url":"https://docs.hopx.ai/cli/commands/usage","quote":"│ Pro          │ $29/mo      │ 100 sandboxes│","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/hopx","score":{"code-sandbox":{"total":56,"rank":49,"of":63},"dev-box":{"total":55,"rank":29,"of":38}},"comparablePrice":{"code-sandbox":29.1456,"dev-box":36.432},"benchmarks":[]},{"slug":"hyperbrowser","name":"Hyperbrowser","company":"Hyperbrowser","website":"https://www.hyperbrowser.ai","docsUrl":"https://www.hyperbrowser.ai/docs/sandboxes/introduction","pricingUrl":"https://www.hyperbrowser.ai/pricing","categories":["browser","code-sandbox"],"tagline":"Hosted browsers for agents at $0.10 per browser-hour, plus Firecracker VM sandboxes billed by vCPU and memory.","summary":"Hyperbrowser sells hosted browser sessions for AI agents and scraping, plus sandboxes: isolated Firecracker VMs for running code, tests and agents. Browser time costs $0.10 per hour plus proxy data, and sandboxes cost $0.045 per vCPU-hour and $0.015 per GiB-hour for allocated resources while running. CPU sandboxes can be saved as memory snapshots, including running processes, and started again from them. All usage draws from one credit balance, with monthly plans that include credits.","isolation":"Firecracker microVM per CPU sandbox; GPU sandboxes use a gVisor runtime","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Node.js","Python","CLI (hx)","REST API"],"regions":"US regions (for example us-west); other regions not stated.","features":{"pauseResume":null,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":256},"pricing":{"rates":{"vcpuHour":0.045,"gibHour":0.015,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not stated for CPU sandboxes and browsers (rates quoted per hour); GPU time per second","billedWhile":"Sandboxes: allocated vCPU and memory for the whole time the sandbox runs, including idle time, until stopped or timed out. Browsers: session time plus proxy data.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.1},"gpu":[{"type":"L4","perHour":0.7992},{"type":"A10G","perHour":1.1016},{"type":"L40S","perHour":1.9512},{"type":"A100 40GB","perHour":2.0988},{"type":"A100 80GB","perHour":2.4984},{"type":"RTX PRO 6000","perHour":3.0312},{"type":"H100","perHour":3.9492},{"type":"H200","perHour":4.5396},{"type":"B200","perHour":6.2496},{"type":"B300","perHour":7.0992}],"plans":[{"name":"Startup","monthlyFee":30,"includedUsageUsd":30,"limits":"30,000 credits ($30) of usage per month across browsers, sandboxes, web data and agents; 25 concurrent sessions; 30-day data retention; residential proxies.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":100,"includedUsageUsd":100,"limits":"100,000 credits ($100) of usage per month; 100 concurrent sessions; 30-day data retention; premium residential proxies.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Free account with 5,000 credits ($5), enough for about 50 browser-hours; 1 concurrent session; upgrade required once used.","notes":["1,000 credits = $1. Browser time is 100 credits ($0.10) per hour; proxy data is $10 per GB on top.","Sandbox memory is billed per allocated GiB. Both rates apply to allocated resources, not actual use.","GPU sandboxes add a per-GPU-second charge (shown per hour) and require contacting Hyperbrowser for access; GPU sandboxes do not support memory snapshots.","Plan credits reset each renewal and do not carry over; separately bought credits expire after 12 months.","Sandbox maximums from the API reference: 32 vCPU and 262,144 MiB (256 GiB) memory, subject to plan and host limits.","Outbound internet is on by default and can be disabled or limited to an allow list. macOS sandboxes are in beta and enabled on request.","AI agent steps (HyperAgent, Browser Use) cost $0.02 per step on top of browser time."]},"bestFor":["Agents that need both a hosted browser and a code sandbox under one credit balance","Cheap browser automation at $0.10 per browser-hour before proxies","Restoring prepared environments quickly from memory snapshots"],"watchOut":["Sandboxes bill allocated vCPU and memory while running, even when idle","Proxy data at $10 per GB can dominate browser costs","GPU and macOS sandboxes need access requests","Free use is a one-time 5,000 credits with a single concurrent session"],"sources":[{"url":"https://www.hyperbrowser.ai/pricing","quote":"|Browser usage |100 credits / hour |$0.10 / hour |","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/pricing","quote":"|vCPU |45 credits / vCPU-hour |$0.045 / vCPU-hour |","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/pricing","quote":"|Memory |15 credits / GiB-hour |$0.015 / GiB-hour |","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/pricing","quote":"Both vCPU and allocated memory are billed while the VM runs.","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/pricing","quote":"Enough for 50 browser hours or 5,000 fetched pages at base rates.","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/pricing","quote":"If you use all the credits on Free, upgrade to continue.","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/pricing","quote":"Your plan's credits reset each time your plan renews, and unused plan credits don't carry over.","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/docs/pricing","quote":"| `H100` | 1.097 | $3.9492 |","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/docs/pricing","quote":"GPU sandboxes are billed per allocated GPU-second, plus the vCPU and system memory rates above.","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/docs/api-reference/create-new-sandbox","quote":"Requests without GPU or macOS options create CPU sandboxes (firecracker).","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/docs/api-reference/create-new-sandbox","quote":"description: Outbound internet access policy.","checked":"2026-11-13"},{"url":"https://www.hyperbrowser.ai/docs/sandboxes/snapshots","quote":"When you take a snapshot, you are freezing the sandbox's memory, processes, and filesystem at that point in time.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/hyperbrowser","score":{"browser":{"total":60,"rank":4,"of":12},"code-sandbox":{"total":62,"rank":34,"of":63}},"comparablePrice":{"browser":0.1,"code-sandbox":30},"benchmarks":[]},{"slug":"instabox","name":"InstaBox","company":"InsForge","website":"https://instabox.sh","docsUrl":"https://instabox.sh/docs/start.md","pricingUrl":"https://instabox.sh/pricing","categories":["dev-box","code-sandbox"],"tagline":"Persistent Firecracker Linux machines for agents, billed per second while running, with plan fees returned as credit.","summary":"InstaBox gives each agent a Firecracker microVM running Debian with Docker, coding agents, a desktop and Chromium in the default template, plus a 20 GiB persistent disk. Machines can be paused with memory kept, stopped, snapshotted, forked and resized, and on paid plans they have no maximum run time by default. vCPU and memory are billed per second only while running; paid plans charge a monthly fee that comes back as usage credit. The free plan runs one 4 vCPU / 8 GiB machine at no charge.","isolation":"Firecracker microVM per machine","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","CLI","REST API","MCP server"],"regions":"Not stated; the API has no region setting.","features":{"pauseResume":true,"snapshots":true,"gpu":true,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.004,"gibHour":0.00225,"instanceHour":null,"instance":null,"storageGibMonth":0.08,"billingIncrement":"per second","billedWhile":"vCPU and memory only while the machine is running (not starting, paused or stopped). On paid plans, persistent disk beyond the plan's included amount is billed while the machine exists, running or not.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"NVIDIA L4 (paid plans; plus the machine's vCPU and memory)","perHour":0.64}],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1 machine (any state) at 4 vCPU / 8 GiB, never billed. 20 GiB fixed disk, 5 snapshots, 200 GiB stored data, 100 GiB egress a month. Pauses after 30 minutes idle; each run starts with a 120-minute limit that grows with use up to 480 minutes. Limited templates; no GPUs.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Pro","monthlyFee":20,"includedUsageUsd":20,"limits":"$20/month returned as usage credit. 100 running machines (500 total), sizes 2/4, 4/8 or 8/16 vCPU/GiB, disk up to 200 GiB per machine, 200 GiB disk included across machines, 1 TiB egress a month, email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Builder","monthlyFee":100,"includedUsageUsd":100,"limits":"$100/month returned as usage credit. 200 running machines (1,000 total), up to 16 vCPU / 32 GiB, disk up to 500 GiB per machine, 1 TiB disk included, 5 TiB egress a month.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":500,"includedUsageUsd":500,"limits":"$500/month returned as usage credit. 500 running machines, unlimited total, up to 32 vCPU / 64 GiB, disk up to 1 TiB per machine, 5 TiB disk included, 50 TiB egress a month, priority support and SLAs.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: one 4 vCPU / 8 GiB machine that is never billed, with a fixed 20 GiB disk, 5 snapshots and 100 GiB egress a month; it pauses after 30 idle minutes, and each run is limited to 120 minutes at first, growing with use up to 480 minutes.","notes":["Paid plan fees come back as monthly usage credit that resets each month; usage beyond it is billed at the unit prices. Credits do not carry over.","The free plan has no dollar allowance: its one machine is simply not billed. The site's estimate cannot model this and lists it only for workloads with no usage.","Disk: free on the free plan. Paid plans include 200 GiB (Pro), 1 TiB (Builder) or 5 TiB (Team) across all machines; beyond that allocated disk costs $0.08 per GiB-month ($0.00011 per GiB-hour), billed while the machine exists. Snapshots, the system disk and a paused machine's saved memory are free.","Egress beyond the plan allowance is $0.02 per GiB. Ingress is free. Outbound bandwidth is capped at about 50 Mbit/s per machine.","GPU: NVIDIA L4 at $0.64 per GPU-hour on paid plans, plus the machine's vCPU and memory; the docs give $0.765 per hour all in per GPU (11 vCPU / 36 GiB per GPU). Max 8 GPUs per organization. An NVIDIA H100 is listed as coming soon at $3.49 per GPU-hour.","macOS machines (Pro and up) lease a whole Mac at one hourly price that includes its vCPUs and memory: M2 $0.45, M2 Pro $0.85, M4 $0.75, M4 Pro $1.25 per Mac-hour, with a 24-hour minimum lease billed whether the machine runs or not.","Max machine size depends on plan: free 4/8, Pro 8/16, Builder 16/32, Team 32/64 (vCPU/GiB). The max shown is Team's.","Paid plans have no maximum run time by default and a configurable idle pause (5-30 minutes or off).","Early accounts (registered before 2026-10-20) keep five free sandboxes until 2026-12-01; after that the free plan runs one at a time.","Outbound HTTPS credentials can be injected by an org-level secrets proxy and ports can be private or public; egress filtering rules are not described in the pages checked.","The company name comes from the SDK package name (@insforge/instabox-sdk); the site does not otherwise name the operator."]},"bestFor":["Persistent per-agent Linux machines with Docker, coding agents and a desktop already installed","Low per-hour compute cost, with small plan fees returned as usage credit","Android app testing with adb and a visible phone screen, on every plan"],"watchOut":["Free plan is one machine, paused after 30 idle minutes, with runs limited to 120 minutes at first and at most 480","Paid plans start at $20/month; there is no fee-free pay-as-you-go option","Disk beyond the plan's included amount is billed even while machines are stopped","Region is not stated and cannot be chosen; outbound bandwidth is capped at about 50 Mbit/s per machine"],"sources":[{"url":"https://instabox.sh/pricing","quote":"| vCPU | $0.004 | per vCPU-hour |","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"| Memory | $0.00225 | per GiB-hour |","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"Runtime is metered by the second, only while a machine is running.","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"beyond that, allocated /data is billed at $0.08 per GiB-month ($0.00011 per GiB-hour). Snapshots are free.","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"**Pro**: $20/month + usage","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"Usage comes out of your included credits first, then you pay as you go. Credits reset every month.","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"| NVIDIA L4 | $0.64 | per GPU-hour |","checked":"2026-11-14"},{"url":"https://instabox.sh/docs/pricing.md","quote":"NVIDIA L4 (`l4`) | \\$0.64 per GPU-hour | 11 vCPU / 36 GiB | \\$0.765 per hour","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"| M2 | $0.45 | $10.80 |","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"| NVIDIA H100 (Coming soon) | $3.49 | per GPU-hour |","checked":"2026-11-14"},{"url":"https://instabox.sh/pricing","quote":"| M4 Pro | $1.25 | $30.00 |","checked":"2026-11-14"},{"url":"https://instabox.sh/docs/billing.md","quote":"A free machine is never billed, so a free month is \\$0.00.","checked":"2026-11-14"},{"url":"https://instabox.sh/docs/limits.md","quote":"Free: 30 minutes, fixed. Paid plans: 5 to 30 minutes, or off","checked":"2026-11-14"},{"url":"https://instabox.sh/docs/limits.md","quote":"Free: each run starts at 120 minutes and grows while the machine is in use, up to 480 minutes","checked":"2026-11-14"},{"url":"https://instabox.sh/docs/limits.md","quote":"| Largest size | 4 vCPU / 8 GiB | 8 vCPU / 16 GiB | 16 vCPU / 32 GiB | 32 vCPU / 64 GiB | 32 vCPU / 64 GiB |","checked":"2026-11-14"},{"url":"https://instabox.sh/llms.txt","quote":"Machine: a Firecracker microVM, Debian 13 (trixie) amd64","checked":"2026-11-14"},{"url":"https://instabox.sh/docs/limits.md","quote":"What a machine sends is capped at about 50 Mbit/s","checked":"2026-11-14"},{"url":"https://instabox.sh/llms.txt","quote":"Sizes (vCPU/GiB): Free 4/8 only; Pro 2/4, 4/8, 8/16; Builder 2/4 up to 16/32; Team 2/4 up to 32/64.","checked":"2026-11-14"},{"url":"https://instabox.sh/llms-full.txt","quote":"A machine has no `region` field in the API, and create does not take one","checked":"2026-11-14"},{"url":"https://instabox.sh/llms.txt","quote":"Every class, method and option of @insforge/instabox-sdk.","checked":"2026-11-14"},{"url":"https://instabox.sh/llms-full.txt","quote":"paid plans only, 8 GPUs per org","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/instabox","score":{"dev-box":{"total":77,"rank":5,"of":38},"code-sandbox":{"total":75,"rank":7,"of":63}},"comparablePrice":{"dev-box":20,"code-sandbox":20},"benchmarks":[]},{"slug":"islo","name":"Islo","company":"Islo","website":"https://islo.dev","docsUrl":"https://docs.islo.dev/overview","pricingUrl":"https://islo.dev/pricing","categories":["dev-box","coding-agent"],"tagline":"Persistent microVM machines for autonomous coding agents, with credential-injecting gateways and per-run cost tracking.","summary":"Islo, from the team behind Incredibuild, runs coding agents on persistent Linux machines, each in its own microVM, with the repository, dependencies, browser and tools prepared in advance. Agents can be triggered from Slack, GitHub or Linear, and a gateway injects credentials and enforces allow or deny rules on outbound traffic. Sandboxes pause on idle or on a runtime cap, keep their files, and can resume on the next request. Plans cost $20 a month for one developer or $200 for a team and include sandbox compute that refreshes weekly; extra time is paid from credits at $0.07 per CPU-hour and $0.04 per GB-hour of provisioned memory.","isolation":"microVM per Islo Machine","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI","REST API"],"regions":"Not stated; the API lists the compute regions allowed for an account. Bring-your-own-cloud deployment is offered.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.07,"gibHour":0.04,"instanceHour":null,"instance":null,"storageGibMonth":0.511,"billingIncrement":"Not stated; rates are quoted per hour","billedWhile":"Only while the sandbox runs, from plan compute or credits: CPU per vCPU, memory per GB provisioned, and disk. No idle fees.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Individual","monthlyFee":20,"includedUsageUsd":0,"limits":"Sandbox compute included (amount not stated), refreshed weekly. Up to 5 sandboxes at once, 1 member. Extra time is paid from credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":200,"includedUsageUsd":0,"limits":"10x the compute of Individual. Up to 50 sandboxes at once, sandboxes up to 32 vCPU and 32 GB memory, unlimited members.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"14-day free trial with a work email and no card; nothing is charged automatically after it.","notes":["Storage is $0.0007 per GB-hour; per 730-hour month that is $0.511 per GB-month. The pricing page says disk is paid only while the machine runs, so a stopped machine may cost less than this rate implies.","Islo states memory in GB without saying GB or GiB; we use the GB-hour rate as the GiB-hour rate.","Bring-your-own-cloud still bills compute time; terms vary by contract.","Islo's docs mention $250 in free credits for users of the Harbor evaluation framework.","How much compute the Individual and Team plans include is not stated; the per-hour rates apply to extra time paid from credits.","Enterprise pricing is custom and adds SSO, admin controls and running Islo in your own cloud."]},"bestFor":["Teams handing tickets from Slack, GitHub or Linear to coding agents that work on a prepared machine","Running agents without giving them raw API keys, since the gateway injects credentials","Agent evaluations that need isolated machines with egress control"],"watchOut":["CPU and memory are billed on provisioned size while running, at higher rates than many sandbox providers","Regions and the amount of compute each plan includes are not published","The product is organized around its own factories, agents and lines, which adds concepts beyond a plain VM"],"sources":[{"url":"https://islo.dev/pricing","quote":"$0.04 /GB-hour","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"$0.0007 /GB-hour","checked":"2026-11-15"},{"url":"https://islo.dev/security/","quote":"microVM per Islo Machine","checked":"2026-11-15"},{"url":"https://islo.dev/","quote":"Built by the team behind Incredibuild","checked":"2026-11-15"},{"url":"https://docs.islo.dev/concepts/sandbox-lifecycle.md","quote":"When a timer fires, the platform pauses the VM: state is snapshotted, compute is released","checked":"2026-11-15"},{"url":"https://docs.islo.dev/llms.txt","quote":"with built-in anti-cheating egress control and $250 in free credits for every Harbor user","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"CPU time $0.07 /CPU-hour per vCPU while the sandbox runs","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"Memory $0.04 /GB-hour per GB you provision","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"Storage $0.0007 /GB-hour per GB of disk","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"Individual ($20 per month) and Team ($200 per month) include sandbox compute.","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"Larger sandboxes, up to 32 vCPU and 32 GB memory","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"14-day free trial · work email · no card required","checked":"2026-11-15"},{"url":"https://islo.dev/pricing","quote":"Billed only while a sandbox runs. No idle fees.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/islo","score":{"dev-box":{"total":57,"rank":28,"of":38},"coding-agent":{"total":82,"rank":1,"of":11}},"comparablePrice":{"dev-box":86,"coding-agent":null},"benchmarks":[]},{"slug":"isorun","name":"Isorun","company":"Isorun","website":"https://isorun.ai","docsUrl":"https://docs.isorun.ai/","pricingUrl":"https://docs.isorun.ai/getting-started/pricing","categories":["code-sandbox","dev-box"],"tagline":"KVM microVMs from any OCI image, billed per second for vCPU and RAM, with hibernate, fork and persistent disks.","summary":"Isorun runs each sandbox as a microVM with its own kernel, booted from any OCI image and driven from a TypeScript SDK, an MCP server or the API. Running sandboxes bill per second at $0.025 per vCPU-hour and $0.015 per GiB-hour, so the default 1 vCPU / 1 GiB size is $0.04 an hour; hibernated sandboxes are not billed for runtime. Hibernate keeps memory, processes and files, and optional persistent disks keep a workspace across sandboxes for $0.02 per GB-month of data stored. The service is in public beta.","isolation":"KVM microVM per sandbox with its own guest kernel","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","MCP server","REST API"],"regions":"US or EU; each organization is scoped to one region chosen at sign-up.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":0.025,"gibHour":0.015,"instanceHour":null,"instance":null,"storageGibMonth":0.02,"billingIncrement":"per second, no minimums","billedWhile":"Wall-clock from create to destroy (or the auto-destroy timer) while running; hibernated sandboxes are not billed for runtime. Persistent disks bill for data stored.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"No plan fee; per-second usage. Concurrency limits are not published.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"$50 of free credit for new accounts, no card required (about 1,250 hours of the default 1 vCPU / 1 GiB sandbox).","notes":["Rates: vCPU $0.025 per hour and RAM $0.015 per GiB-hour. Listed shapes: 1 vCPU + 1 GiB at $0.040/h (default), 2 vCPU + 4 GiB at $0.110/h, 4 vCPU + 8 GiB at $0.220/h, 8 vCPU + 16 GiB at $0.440/h. The maximum size we show is the largest listed shape; no maximum is stated.","Unit wording differs: the homepage says '1 vCPU + 1 GB RAM' while the docs price RAM per GiB. We use the docs' GiB rate.","Persistent disks cost $0.02 per GB-month for data actually stored, not provisioned size, plus normal compute while a disk is attached. This is the storage rate we show. No separate charge is stated for hibernation snapshots; the docs say a hibernated sandbox costs nothing on the runner side.","The homepage says egress traffic and platform fees are included in the compute price.","Every sandbox has an idle timeout (default 300 seconds) after which it is destroyed, not hibernated; it is configurable per create and 0 disables it. Destroy deletes the sandbox and any hibernation snapshot.","Default sandbox: 1 vCPU, 1 GiB RAM and a 4 GiB writable scratch disk that grows on write. /workspace is ephemeral unless a persistent disk is attached.","How long a hibernated sandbox is kept is not stated.","Public beta. SOC 2 Type II is listed as coming soon. No GPU support is described."]},"bestFor":["Short code-execution runs from any Docker image with per-second billing and no minimum.","Agent workspaces that hibernate between sessions and resume with processes and memory intact.","Locking down egress with allow-lists and keeping API keys out of the sandbox."],"watchOut":["Public beta.","Sandboxes are destroyed after 5 idle minutes by default unless the timeout is changed or the sandbox is hibernated.","Only a TypeScript SDK is offered; other languages use the API or MCP server.","Retention of hibernated sandboxes is not documented."],"sources":[{"url":"https://docs.isorun.ai/getting-started/pricing","quote":"vCPU |$0.025 / hour","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/getting-started/pricing","quote":"RAM |$0.015 / GiB / hour","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/getting-started/pricing","quote":"8 vCPU + 16 GiB |$0.440 / hour","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/getting-started/pricing","quote":"Every new account gets $50 in free credit, about 1,250 hours of a default sandbox. No card required to start.","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/getting-started/pricing","quote":"Hibernated sandboxes: not billed for active runtime; you pay nothing while the sandbox is paused.","checked":"2026-11-14"},{"url":"https://isorun.ai","quote":"Public beta, $50 free credit.","checked":"2026-11-14"},{"url":"https://isorun.ai","quote":"Billed per second at $0.04 /hour for 1 vCPU + 1 GB RAM.","checked":"2026-11-14"},{"url":"https://isorun.ai","quote":"Simple pricing: pay for compute, egress traffic and platform fees are included.","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/sandboxes/persistent-storage","quote":"$0.02 per GB-month for data stored — you pay for actual data, not provisioned size — plus the usual per-second compute while a disk is attached.","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/sandboxes/lifecycle","quote":"Every sandbox has an idle timeout (default 300 seconds, configurable per-create via `timeoutSec` ). When the timer fires, the sandbox is destroyed.","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/sandboxes/lifecycle","quote":"It boots a microVM with its own kernel.","checked":"2026-11-14"},{"url":"https://docs.isorun.ai/getting-started/authentication","quote":"the organization is scoped to a region (US or EU)","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/isorun","score":{"code-sandbox":{"total":78,"rank":1,"of":63},"dev-box":{"total":79,"rank":2,"of":38}},"comparablePrice":{"code-sandbox":19.36,"dev-box":24.2},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":96,"unit":"ms","better":"lower","date":"2026-11-13","rank":1,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":36034,"unit":"ms","better":"lower","date":"2026-11-13","rank":2,"of":25}]},{"slug":"judge0","name":"Judge0","company":"Judge0 d.o.o.","website":"https://judge0.com","docsUrl":"https://ce.judge0.com/","pricingUrl":"https://judge0.com","categories":["code-sandbox"],"tagline":"Open-source code execution API for 60+ languages, self-hosted for free or used as a cloud service priced per submission.","summary":"Judge0 is a GPL-3.0 code execution system that compiles and runs submissions in 60+ languages inside a sandbox with configurable time and memory limits. You can self-host it with Docker at no cost or use the hosted cloud API. The cloud is priced per submission, not per hour: a free Basic plan allows 50 submissions a day, daily-quota plans start at $33.75 a month, with $0.0013 for each submission over the daily quota. It suits running single programs, not long-lived environments.","isolation":"Each submission runs in the isolate sandbox (Linux namespaces and control groups) inside Judge0's containers","isolationKind":"process","openSource":true,"selfHost":true,"license":"GPL-3.0","repoUrl":"https://github.com/judge0/judge0","lastRelease":"2026-10-26","sdks":["REST API","Python"],"regions":"Not stated","features":{"pauseResume":false,"snapshots":false,"gpu":false,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per submission (cloud); self-hosted is free software","billedWhile":"Cloud: each submission counts against a daily quota or is billed per submission; run time is not metered. Self-hosted: you pay only for your own servers.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Basic (cloud)","monthlyFee":0,"includedUsageUsd":0,"limits":"50 submissions per day at no cost; no credit card required.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Pro (cloud)","monthlyFee":33.75,"includedUsageUsd":0,"limits":"2,000 submissions per day; $0.0013 per extra submission; community and email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Ultra (cloud)","monthlyFee":67.5,"includedUsageUsd":0,"limits":"5,000 submissions per day; $0.0013 per extra submission; community and email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Mega (cloud)","monthlyFee":133.75,"includedUsageUsd":0,"limits":"10,000 submissions per day; $0.0013 per extra submission; community and email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Self-hosting the open-source version is free with no limits; the hosted Basic plan includes 50 submissions per day at no cost.","notes":["There is no hourly compute price, so hourly rates are left empty. Cloud plans are priced per submission or by daily submission quota.","judge0.com shows prices in euros by default: Pro €27, Ultra €54 and Mega €107 a month, and €0.001 per extra submission. The US dollar prices used here appear when you switch the currency to USD, which shows 1.25 times the euro price.","Dedicated Cloud pricing was not captured.","Each submission can set its own time and memory limits; platform maximums are not stated on the pages checked.","The last tagged release is v1.13.1 (April 2024); the repository had commits as recent as 26 October 2026. Version 1.13.1 fixed three critical security vulnerabilities found in versions up to 1.13.0.","A pay-per-use cloud plan is also offered. Its price appears only in an interactive tab on the pricing page, so it is not listed here."]},"bestFor":["Coding-assessment, judge and e-learning platforms that run single programs in many languages","Teams that want to self-host a code execution API at no license cost","Low-volume use on the free 50-submissions-per-day cloud plan"],"watchOut":["Built for one-shot program runs, not persistent or interactive environments","Process-level sandbox rather than a VM boundary; versions up to 1.13.0 had critical security flaws","Last tagged release dates from 2024","GPL-3.0 license applies if you modify and distribute it"],"sources":[{"url":"https://judge0.com","quote":"Yes! Judge0 offers multiple free options: you can self-host the open-source version with no limitations, or use our Basic cloud plan which includes 50 submissions per day at no cost.","checked":"2026-11-13"},{"url":"https://judge0.com","quote":"€27/month","checked":"2026-11-14"},{"url":"https://judge0.com","quote":"€0.001 per extra submission","checked":"2026-11-14"},{"url":"https://judge0.com","quote":"10 000 submissions/day","checked":"2026-11-13"},{"url":"https://judge0.com","quote":"Support for 60+ programming languages","checked":"2026-11-13"},{"url":"https://judge0.com","quote":"Each submission runs in a secure, containerized environment with strict resource limits and network restrictions.","checked":"2026-11-13"},{"url":"https://github.com/judge0/judge0","quote":"Robust, fast, scalable, and sandboxed open-source online code execution system for humans and AI.","checked":"2026-11-13"},{"url":"https://github.com/judge0/judge0/blob/master/CHANGELOG.md","quote":"uses updated Isolate","checked":"2026-11-13"},{"url":"https://github.com/judge0/judge0/blob/master/CHANGELOG.md","quote":"found three critical vulnerabilities in Judge0 versions `<= 1.13.0`, which this release fixes","checked":"2026-11-13"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/judge0","score":{"code-sandbox":{"total":46,"rank":61,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"kata-containers","name":"Kata Containers","company":"Kata Containers community (OpenInfra)","website":"https://katacontainers.io","docsUrl":"https://github.com/kata-containers/kata-containers/tree/main/docs","pricingUrl":"https://github.com/kata-containers/kata-containers","categories":["code-sandbox"],"tagline":"Building block: open-source container runtime that runs each container or pod inside a lightweight VM.","summary":"Kata Containers is a building block, not a ready-made sandbox service: a container runtime that starts each container or Kubernetes pod in its own lightweight virtual machine, so workloads get VM isolation while keeping container tooling. It supports several hypervisors and runs on x86_64, ARM64, IBM Power and IBM Z. Kubernetes sandbox projects use it as a runtime class. It is free Apache-2.0 software.","isolation":"Lightweight VM per container or pod, using a choice of hypervisors with hardware virtualization","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/kata-containers/kata-containers","lastRelease":"2026-10-21","sdks":["OCI runtime","containerd shim","Kubernetes RuntimeClass"],"regions":"Runs on your own 64-bit hosts with hardware virtualization (x86_64, ARM64, ppc64le, s390x).","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own hardware","billedWhile":"Not applicable: no usage charges","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license.","notes":["A runtime layer: it runs standard container images but leaves APIs, scheduling and sandbox lifecycle to Kubernetes or your own platform.","The hypervisor is chosen in configuration; see the project's hypervisors document."]},"bestFor":["Giving Kubernetes pods VM-level isolation through a runtime class","Running untrusted container workloads with existing container tooling","Platforms that need IBM Power or IBM Z support"],"watchOut":["Building block only: no sandbox API or SDK of its own","Requires hardware virtualization on the host (or nested virtualization in cloud VMs)","More moving parts than a plain container runtime: hypervisor, guest kernel and agent"],"sources":[{"url":"https://github.com/kata-containers/kata-containers","quote":"Kata Containers is an open source project and community working to build a standard implementation of lightweight Virtual Machines (VMs) that feel and perform like containers, but provide the workload isolation and security advantages of VMs.","checked":"2026-11-13"},{"url":"https://github.com/kata-containers/kata-containers","quote":"The code is licensed under the Apache 2.0 license.","checked":"2026-11-13"},{"url":"https://github.com/kata-containers/kata-containers","quote":"Kata Containers currently runs on 64-bit systems supporting the following technologies","checked":"2026-11-13"},{"url":"https://github.com/kata-containers/kata-containers/releases/tag/4.3.1","quote":"4.3.1","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/kata-containers","score":{"code-sandbox":{"total":58,"rank":43,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"kernel","name":"Kernel","company":"Kernel","website":"https://www.kernel.sh","docsUrl":"https://www.kernel.sh/docs","pricingUrl":"https://www.kernel.sh/pricing","categories":["browser"],"tagline":"Cloud Chromium browsers in per-browser unikernel microVMs, billed per second by memory and paused when idle.","summary":"Kernel runs each Chromium browser in its own unikernel microVM, driven with Playwright, CDP or computer controls. Usage is billed per second at $0.0000166667 per GB-second: a 1 GB headless browser costs about $0.06 per hour and the default 8 GB headful browser about $0.48 per hour. Browsers go into standby about 10 seconds after the last client disconnects, which stops usage charges, and proxies are not charged. Every plan includes monthly usage credit, starting with $5 on the free plan.","isolation":"Each browser runs in its own unikernel microVM with its own kernel and filesystem, isolated at the hypervisor","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/kernel/kernel-images","lastRelease":"2026-11-10","sdks":["TypeScript","Python","Go","CLI","MCP server","Playwright","CDP"],"regions":"US by default; regional browsers are available on the Start-Up and Enterprise plans at the same rates.","features":{"pauseResume":true,"snapshots":true,"gpu":true,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":null,"gibHour":0.06000012,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Per second","billedWhile":"Only while the browser is active; standby (about 10 seconds after the last client disconnects) stops charges, and idle browsers in a pool are not charged","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.06000012},"gpu":[{"type":"Headful browser with GPU acceleration","perHour":2.88000576}],"plans":[{"name":"Developer","monthlyFee":0,"includedUsageUsd":5,"limits":"$5 of usage credit per month; 5 concurrent browsers; stealth mode with captcha solver and proxy; 3 managed auth connections; 1-day replays; no configurable or custom proxies.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Hobbyist","monthlyFee":30,"includedUsageUsd":10,"limits":"$10 of usage credit per month; 10 concurrent browsers; unlimited proxy usage; 7-day replays; unlimited managed auth connections and vaults.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Start-Up","monthlyFee":200,"includedUsageUsd":50,"limits":"$50 of usage credit per month; 150 concurrent browsers; bring-your-own proxies; GPU acceleration; regional browsers; 30-day replays.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Developer plan: $5 of usage credit every month, 5 concurrent browsers, no credit card required.","notes":["Usage rates per second: headless browser $0.0000166667, headful browser $0.0001333336, headful with GPU $0.0008000016. Multiplied by 3,600 that is $0.06000012, $0.48000096 and $2.88000576 per hour.","The pricing page states the rate as $0.0000166667 per GB-second ($0.06000012 per GB-hour; we treat GB as GiB, which Kernel does not specify). Headless browsers come in 1, 2 or 4 GB and headful in 8 GB (default) or 16 GB; the per-hour browser price shown is the documented headless rate, which matches 1 GB.","Proxies are not charged on any plan. Plan credits cover usage only, not the monthly fee.","Browsers in standby keep their state and stop usage charges but still count toward the concurrency limit. A browser is deleted after its timeout (default 60 seconds, up to 72 hours) once in standby.","App invocations (code deployed to run next to the browsers) are billed at $0.0000666668 per second.","GPU acceleration is only on Start-Up and Enterprise. Enterprise pricing is custom.","The browser images are open source (Apache-2.0) and can run in Docker or on Unikraft unikernels; the hosted platform features are not part of that repository."]},"bestFor":["Agents that wait on users or models between actions, since standby stops charges","Untrusted pages or code next to the browser, given per-browser VM isolation","Proxy-heavy work, since proxy traffic is not metered"],"watchOut":["Headful browsers cost eight times the headless rate because billing follows memory","Standby browsers still take a concurrency slot until deleted","GPU, custom proxies and regional browsers need the $200/month Start-Up plan","Only 5 concurrent browsers on the free plan"],"sources":[{"url":"https://www.kernel.sh/pricing","quote":"Usage costs: $0.0000166667 per GB-second (no charges for idle time or proxies).","checked":"2026-11-13"},{"url":"https://www.kernel.sh/pricing","quote":"$5 / month free credits","checked":"2026-11-13"},{"url":"https://www.kernel.sh/docs/info/pricing","quote":"Headless browser |0.0000166667","checked":"2026-11-13"},{"url":"https://www.kernel.sh/docs/info/pricing","quote":"Headful browser + GPU acceleration |0.0008000016","checked":"2026-11-13"},{"url":"https://www.kernel.sh/docs/info/pricing","quote":"Start-Up |$200 + usage |$50","checked":"2026-11-13"},{"url":"https://www.kernel.sh/docs/browsers/concurrency-and-limits","quote":"Memory |8 GB or 16 GB |1 GB, 2 GB, or 4 GB","checked":"2026-11-13"},{"url":"https://www.kernel.sh/docs/overview/why-kernel","quote":"Each browser is a microVM with its own kernel and filesystem","checked":"2026-11-13"},{"url":"https://www.kernel.sh/docs/browsers/standby","quote":"Kernel browsers automatically start entering standby after ten seconds with no activity.","checked":"2026-11-13"},{"url":"https://github.com/kernel/kernel-images","quote":"This image can be used to run headful Chromium in a Docker container or with Unikraft unikernels.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/kernel","score":{"browser":{"total":76,"rank":3,"of":12}},"comparablePrice":{"browser":0.06000012},"benchmarks":[]},{"slug":"kosa8","name":"kosa8 Cloud","company":"Kosa8 Labs","website":"https://kosa8.com","docsUrl":"https://docs.kosa8.com/guides/cloud/","pricingUrl":"https://docs.kosa8.com/guides/cloud/","categories":["code-sandbox"],"tagline":"Hosted microVM sandboxes from the kosa8 agent-security tool, billed per second and asleep for snapshot cost only.","summary":"kosa8 runs each agent in a microVM with its own kernel, either on your own Mac or Linux machines or on kosa8 Cloud, its hosted service in the US (Dallas, Texas) and the EU (Helsinki). Cloud sandboxes are created from Python, JavaScript/TypeScript, Go, Ruby or Java SDKs or a REST API and bill per second at $0.048 per vCPU-hour and $0.0154 per GB-hour of RAM. Idle sandboxes go to sleep and then cost only $0.08 per GB-month for their snapshot. Cloud is pay-as-you-go with a card on file and no free credit; the paid kosa8 plans ($29 to $199 a month) add features for your own machines and include no Cloud credit.","isolation":"Hardware-virtualised microVM per sandbox with its own kernel (Cloud Hypervisor on KVM on Linux)","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","Go","Ruby","Java","REST API","MCP server","CLI"],"regions":"US (us-ash1, Dallas, Texas) and EU (eu-hel1, Helsinki), at the same price. A sandbox goes to the region nearest the caller unless you name one or set a default region for the organization. Asia-Pacific is planned at 1.2x the price. You can also join your own servers as hosts.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.048,"gibHour":0.0154,"instanceHour":null,"instance":null,"storageGibMonth":0.08,"billingIncrement":"per second","billedWhile":"While a sandbox runs. An idle sandbox sleeps on a timer you set and then pays only for its snapshot until the next command wakes it.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Cloud pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"No plan needed; pay only for Cloud usage. A payment method is required before anything runs. 20 sandboxes running or paused per organization (sleeping ones do not count).","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Solo","monthlyFee":29,"includedUsageUsd":0,"limits":"4 forks at once, 64 GB sandboxes; overage $0.10 per fork-hour if turned on.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":89,"includedUsageUsd":0,"limits":"16 forks at once, 128 GB sandboxes.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":169,"includedUsageUsd":0,"limits":"Per person per month, from 3 people. 48 shared forks at once, 256 GB sandboxes, governance and MCP policies.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":199,"includedUsageUsd":0,"limits":"Per person per month, from 5 people. 160 shared forks at once, 500 GB sandboxes.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No Cloud credit of any kind: a card is required and usage is billed per second from the first second. Until its first payment, a new organization can run up to $50 of usage a month. Local sandboxes on your own computer are free.","notes":["Plans are charged at checkout with no trial period and include no Cloud credit. The kosa8 FAQ still says plans include $10 to $50 a month of Cloud credit, but the pricing page and Cloud guide say they do not; we follow the pricing page. From 1 December 2026, Cloud usage is billed on top of any plan.","Cloud rates are the same in the US (Dallas) and EU (Helsinki) regions; Asia-Pacific will cost 1.2x when it opens.","Sleeping sandboxes pay $0.08 per GB-month for their snapshot. Bring-your-own hosts cost $0.20 per host-hour while registered, with no compute charge.","Plan limits (forks at once, sandbox size in GB) apply to kosa8's fork feature and local use; ordinary sandboxes are not counted. Team and Scale fees and credits are per person.","Cloud templates today are base (Alpine), python and node; custom images on Cloud and egress controls on Cloud are not documented (the FAQ says the egress broker is still coming to Cloud).","Maximum vCPU and memory per Cloud sandbox are not stated.","After sign-up with a card on file, sandboxes, save and undo and the dashboard are free on your own computer; forks, local models and the other features need a plan."]},"bestFor":["Teams that want the same sandboxes on laptops, their own servers and a hosted cloud","Agents that idle between commands, since sleeping sandboxes pay only snapshot storage","EU data residency on the hosted service, by setting the organization's default region to Helsinki"],"watchOut":["Hosted sandboxes go to the nearest region, US or EU, unless you set a default region for the organization","A card is required before any Cloud sandbox runs","Maximum sandbox size on Cloud is not published","Several security features shown on the site (egress broker, undo) apply to local use and are not yet on Cloud","Plans add no Cloud credit; all Cloud usage is billed"],"sources":[{"url":"https://docs.kosa8.com/guides/cloud/","quote":"Billed per second while a sandbox runs: $0.048 per vCPU-hour","checked":"2026-11-15"},{"url":"https://docs.kosa8.com/guides/cloud/","quote":"An idle sandbox goes to sleep and then costs only $0.08 per GB-month for its snapshot","checked":"2026-11-15"},{"url":"https://docs.kosa8.com/guides/cloud/","quote":"The organization already has 20 sandboxes running or paused (sleeping ones do not count, but waking one counts it again).","checked":"2026-11-15"},{"url":"https://kosa8.com/pricing/","quote":"Plans are charged at checkout; there is no trial period.","checked":"2026-11-15"},{"url":"https://kosa8.com/pricing/","quote":"$169 /person /month","checked":"2026-11-15"},{"url":"https://kosa8.com/faq/","quote":"each sandbox is hardware-virtualised with its own kernel (Apple's Virtualization framework on a Mac, Cloud Hypervisor on KVM on Linux)","checked":"2026-11-15"},{"url":"https://kosa8.com/faq/","quote":"kosa8 Cloud runs in the US (us-ash1, Dallas, Texas) and the EU (eu-hel1, Helsinki).","checked":"2026-11-15"},{"url":"https://kosa8.com/faq/","quote":"To keep every sandbox in the EU, set your organisation's default region to eu-hel1 in Settings.","checked":"2026-11-15"},{"url":"https://docs.kosa8.com/guides/cloud/","quote":"in the US and the EU, the same in both; Asia-Pacific will cost 1.2× when it opens.","checked":"2026-11-15"},{"url":"https://docs.kosa8.com/guides/cloud/","quote":"kosa8 Cloud gives no credit: no plan includes Cloud credit, and there is no sign-up or referral credit.","checked":"2026-11-15"},{"url":"https://docs.kosa8.com/guides/cloud/","quote":"There is no credit of any kind: usage is billed per second to the card on file, from the first second.","checked":"2026-11-15"},{"url":"https://kosa8.com/pricing/","quote":"Plans do not include kosa8 Cloud credit.","checked":"2026-11-15"},{"url":"https://kosa8.com/pricing/","quote":"Until its first payment, a self-serve account can run up to $50 of usage a month.","checked":"2026-11-15"},{"url":"https://kosa8.com/faq/","quote":"Do sandboxes have a GPU? Not on kosa8 Cloud.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/kosa8","score":{"code-sandbox":{"total":58,"rank":44,"of":63}},"comparablePrice":{"code-sandbox":27.7376},"benchmarks":[]},{"slug":"koyeb-sandboxes","name":"Koyeb Sandboxes","company":"Koyeb","website":"https://www.koyeb.com/docs/sandboxes","docsUrl":"https://www.koyeb.com/docs/sandboxes","pricingUrl":"https://www.koyeb.com/docs/reference/instances","categories":["code-sandbox","gpu"],"tagline":"MicroVM sandboxes on Koyeb's serverless platform, billed per second at Koyeb instance prices, with CPU or GPU.","summary":"Koyeb Sandboxes are ephemeral environments created from Docker images with Koyeb's Python and JavaScript SDKs, running on the same microVM instances as Koyeb services. Each sandbox is billed per second at the price of its instance type, from $0.0036 per hour for 0.25 vCPU / 256 MB up to GPU instances such as an H100 at $2.50 per hour. Sandboxes scale to zero after an idle timeout and can be auto-deleted after up to 24 hours, and filesystem or full snapshots can seed new sandboxes. Sandboxes are in public preview and need a Starter (pay-as-you-go), Pro or Scale plan.","isolation":"MicroVM per instance on bare-metal servers","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript"],"regions":"Koyeb regions in the US, EU and Asia (default 'na'); GPU instances only in specific regions.","features":{"pauseResume":true,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":40,"maxMemoryGib":128},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.0144,"instance":"small: 1 vCPU / 1 GB / 10 GB SSD","storageGibMonth":null,"billingIncrement":"per second","billedWhile":"Per second at the instance type's price while the sandbox's instance runs; after the idle timeout (default 300 seconds) it scales to zero and compute stops.","cpuActiveOnly":false,"unitVcpu":1,"unitGib":1,"browserHour":null},"gpu":[{"type":"NVIDIA RTX 4000 SFF Ada 20 GB","perHour":0.5},{"type":"NVIDIA L4 24 GB","perHour":0.7},{"type":"NVIDIA RTX A6000 48 GB","perHour":0.75},{"type":"NVIDIA L40S 48 GB","perHour":1.2},{"type":"NVIDIA A100 80 GB","perHour":1.6},{"type":"NVIDIA RTX PRO 6000 96 GB","perHour":2.2},{"type":"NVIDIA H100 80 GB","perHour":2.5},{"type":"NVIDIA H200 141 GB","perHour":3},{"type":"NVIDIA B200 180 GB","perHour":5.5},{"type":"2x NVIDIA H100 80 GB","perHour":5},{"type":"8x NVIDIA H100 80 GB","perHour":20}],"plans":[{"name":"Starter","monthlyFee":0,"includedUsageUsd":0,"limits":"Pay as you go with a valid payment method; no base fee.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":29,"includedUsageUsd":10,"limits":"$10 of compute included each month. 10 users, 100 services, CPU concurrency 100, GPU concurrency 20, 1 TB/month bandwidth.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":299,"includedUsageUsd":100,"limits":"$100 of compute included each month. 50 users, 1,000 services, CPU concurrency 1,000, GPU concurrency 40, AWS regions, 99.9% uptime SLA.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free sandbox usage is stated. Koyeb's one free instance is for a single web service and is not offered for sandboxes.","notes":["Sandboxes are priced as Koyeb instances. Standard sizes from small up are linear at $0.0144 per hour per 1 vCPU / 1 GB unit, using whichever of vCPU or memory needs more units (for example 3xlarge, 24 vCPU / 32 GB, is $0.4608). Smaller sizes are cheaper: nano 0.25 vCPU / 256 MB $0.0036, micro 0.5 vCPU / 512 MB $0.0072 (the SDK default).","Eco instances (Washington, D.C., Frankfurt, Singapore) cost the same per hour for half the vCPU of the matching standard size; whether sandboxes accept Eco types is not stated.","Koyeb states memory in GB; we treat it as GiB.","GPU instance prices include their vCPU and RAM (for example H100: 15 vCPU, 180 GB RAM). Koyeb says sandboxes are available on both CPU and GPU instances.","Auto-deletion can be set from 60 seconds to 24 hours after creation, or up to 12 hours after scaling to zero; idle_timeout=0 keeps a sandbox always on.","Snapshots can be filesystem-only or full and are used to create new sandboxes. Outbound traffic can be blocked or limited to an IP/CIDR allowlist.","Bandwidth beyond the included 1 TB/month costs $0.02/GB in EU/US and $0.04/GB in Asia. Snapshot and volume storage prices are not stated on the pages checked.","The largest standard instance is 5xlarge (40 vCPU / 128 GB). Koyeb sandboxes are in public preview."]},"bestFor":["Sandboxes that need a GPU, from an L4 to multi-GPU H100 and H200 machines","Teams already deploying services on Koyeb","Bursty sandboxes that should scale to zero between requests"],"watchOut":["Still in public preview","No free sandbox usage; a payment method or paid plan is required","Auto-deletion is capped at 24 hours after creation","Storage pricing for snapshots is not published on the pages checked"],"sources":[{"url":"https://www.koyeb.com/docs/reference/instances","quote":"|`small` |1 |1GB |10GB SSD |$0.000004 |$0.0144 |$10.71 |","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/reference/instances","quote":"Under the hood, Instances are microVMs which serve as a thin, isolated, high performance wrapper around your code.","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/reference/instances","quote":"|`gpu-nvidia-h100` |80 GB |15 |180 GB |$0.000694 |$2.50 |$1860 |","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/sandboxes","quote":"Sandboxes are available for users on the Starter, Pro, or Scale plans.","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/reference/organizations","quote":"The Pro plan has a fixed cost of $29 per month. This includes $10 per month of compute resources usage","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/reference/organizations","quote":"Available to all users with a valid payment method. The Starter plan has no baseline cost","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/sandboxes/sandbox-lifecycle","quote":"For all plans (Starter, Pro, and Scale), auto-deletion can be configured with a minimum delay of","checked":"2026-11-13"},{"url":"https://www.koyeb.com/docs/sandboxes/sandbox-lifecycle","quote":"Koyeb Sandboxes are currently in public preview.","checked":"2026-11-13"},{"url":"https://github.com/koyeb/koyeb-python-sdk/blob/main/docs/sandbox.md","quote":"`instance_type` - Instance type (default: micro)","checked":"2026-11-13"},{"url":"https://github.com/koyeb/koyeb-python-sdk/blob/main/docs/sandbox.md","quote":"`outbound_allowlist` - List of IPs/CIDRs allowed as outbound destinations;","checked":"2026-11-13"},{"url":"https://www.koyeb.com/blog/koyeb-sandboxes-javascript-sdk-now-available","quote":"Each sandbox runs with true isolation, is available on both CPU and GPU Instances","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/koyeb-sandboxes","score":{"code-sandbox":{"total":76,"rank":3,"of":63},"gpu":{"total":69,"rank":4,"of":7}},"comparablePrice":{"code-sandbox":10.137599999999999,"gpu":2.5},"benchmarks":[]},{"slug":"kubernetes-agent-sandbox","name":"Kubernetes Agent Sandbox","company":"Kubernetes SIG Apps","website":"https://agent-sandbox.sigs.k8s.io","docsUrl":"https://agent-sandbox.sigs.k8s.io/docs/","pricingUrl":"https://github.com/kubernetes-sigs/agent-sandbox","categories":["code-sandbox"],"tagline":"Kubernetes controller for stateful agent sandboxes with warm pools; offered on GKE at no extra charge.","summary":"Agent Sandbox is a Kubernetes SIG Apps project that adds a Sandbox custom resource and controller for long-running, stateful single-pod workloads with a stable identity and persistent storage. Extensions add templates, claims and pre-warmed pools for fast allocation, plus Go and Python SDKs and an optional router. It orchestrates pods and leaves kernel isolation to a runtime class such as gVisor or Kata Containers. Google offers it on GKE at no extra charge beyond normal GKE resource pricing.","isolation":"Kubernetes pod per sandbox; stronger isolation comes from a RuntimeClass such as gVisor or Kata Containers (GKE uses gVisor or microVMs)","isolationKind":"container","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/kubernetes-sigs/agent-sandbox","lastRelease":"2026-11-12","sdks":["Go","Python","Kubernetes API (kubectl/YAML)"],"regions":"Any Kubernetes 1.25+ cluster you run (1.29+ recommended), including GKE.","features":{"pauseResume":true,"snapshots":null,"gpu":true,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own cluster","billedWhile":"Not applicable: you pay only for the cluster resources the pods use","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under Apache-2.0. On GKE it is offered at no extra charge; normal GKE pricing applies to the resources you create.","notes":["Isolation strength depends on the runtime class you configure. Google's GKE docs warn that without gVisor or microVMs the pods run without kernel isolation and should not run untrusted code.","On GKE, Pod Snapshots can suspend idle sandboxes and resume them; that is a GKE feature, not part of the open-source controller.","GKE docs list GPUs and TPUs as a reason to choose Agent Sandbox; any pod resource your cluster supports can be requested."]},"bestFor":["Teams already running Kubernetes that want a standard API for per-agent sandboxes","Reinforcement learning and evaluation loops that need fast claims from warm pools","GKE users who want managed sandbox support at no extra charge"],"watchOut":["Plain pods share the host kernel; you must configure gVisor, Kata or another runtime for untrusted code","You operate the cluster, controller upgrades and networking yourself","Requires Kubernetes 1.25 or later with CEL validation for CRDs"],"sources":[{"url":"https://github.com/kubernetes-sigs/agent-sandbox","quote":"It delegates low-level container isolation to secure \"Sandbox Runtimes\" (like gVisor or Kata Containers)","checked":"2026-11-13"},{"url":"https://github.com/kubernetes-sigs/agent-sandbox","quote":"The Sandbox controller manages the lifecycle of the pod, including creation, scheduled deletion, pausing and resuming.","checked":"2026-11-13"},{"url":"https://github.com/kubernetes-sigs/agent-sandbox","quote":"Agent Sandbox requires Kubernetes 1.25 or later","checked":"2026-11-13"},{"url":"https://docs.cloud.google.com/kubernetes-engine/docs/how-to/agent-sandbox","quote":"Agent Sandbox on GKE is open source software that you can use at no extra charge.","checked":"2026-11-13"},{"url":"https://docs.cloud.google.com/kubernetes-engine/docs/how-to/how-install-agent-sandbox","quote":"If you don't specify a GKE Sandbox runtime (gVisor or microVMs), your Pods run without kernel isolation, so don't use them to run untrusted code.","checked":"2026-11-13"},{"url":"https://docs.cloud.google.com/kubernetes-engine/docs/how-to/how-install-agent-sandbox","quote":"Use Agent Sandbox on GKE when each agent needs a dedicated Pod, GPUs or TPUs, a full Kubernetes PodSpec, or a GKE Autopilot cluster.","checked":"2026-11-13"},{"url":"https://github.com/kubernetes-sigs/agent-sandbox/releases/tag/v1.0.11","quote":"v1.0.11","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/kubernetes-agent-sandbox","score":{"code-sandbox":{"total":54,"rank":54,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"langsmith-sandboxes","name":"LangSmith Sandboxes","company":"LangChain","website":"https://www.langchain.com/langsmith/sandboxes","docsUrl":"https://docs.langchain.com/langsmith/sandboxes","pricingUrl":"https://www.langchain.com/pricing-langsmith","categories":["code-sandbox"],"tagline":"MicroVM sandboxes inside LangSmith, billed per second for vCPU, memory and storage, with a small monthly allowance.","summary":"LangSmith Sandboxes are microVMs for agent-generated code, created from the LangSmith Python or TypeScript SDK or a CLI. Agents get root, Docker and a default Ubuntu image with common languages and tools, or you boot from your own container-built snapshot. Compute, memory and storage are billed per second, and every plan includes 8 LSU (US$8) of sandbox usage each month. Sandboxes stop after an idle timeout, wake on the next command, and can be snapshotted and forked.","isolation":"MicroVM per sandbox","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI"],"regions":"LangSmith Cloud regions (US or EU); also available on BYOC and self-hosted deployments.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.0576,"gibHour":0.01845,"instanceHour":null,"instance":null,"storageGibMonth":0.08979,"billingIncrement":"per second","billedWhile":"Compute and memory while the sandbox runs; storage for persistent sandboxes while their filesystem is kept. Stopped sandboxes pay storage only.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Developer","monthlyFee":0,"includedUsageUsd":8,"limits":"1 seat. Includes 8 LSU (US$8) of sandbox usage per month; limited to 10 sandboxes.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Plus","monthlyFee":39,"includedUsageUsd":8,"limits":"$39 per seat per month. Includes 8 LSU (US$8) of sandbox usage per month; usage beyond that is pay as you go.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Every plan, including the free Developer plan, includes 8 LSU (US$8) of sandbox usage each month. Developer is limited to 10 sandboxes.","notes":["Prices are in LangChain Standard Units (LSU); 1 LSU = $1. Rates: 0.0576 LSU per vCPU-hour, 0.01845 LSU per GiB-hour of memory, 0.000123 LSU per GiB-hour of storage.","Storage is shown per GiB-month as 0.000123 x 730 hours.","The pages checked do not say whether CPU is billed on the requested size or on actual use; sandboxes can burst to twice their requested CPU when the host has spare capacity.","Sizes: 0.05 to 16 vCPU (default 0.5) and up to 64 GiB memory (default 4 GiB per vCPU, within 50% of that ratio).","Sandboxes stop after 10 minutes idle by default (configurable, or disabled) and are deleted 14 days after stopping by default (up to 30 days). Non-persistent sandboxes discard writes on stop and have no storage charge.","Non-Enterprise organizations get restricted egress to a managed allowlist of package registries, source hosts and model APIs; unrestricted egress needs a support request.","The Plus plan's seat fee covers all of LangSmith, not just sandboxes. Self-hosting is part of the Enterprise plan."]},"bestFor":["Teams already using LangSmith or Deep Agents that want sandboxes in the same account","Agents that need root, Docker and a full toolchain in a disposable VM","Branching agent runs by snapshotting a sandbox and forking it"],"watchOut":["Outbound network access is restricted to an allowlist unless you request an exemption","The free Developer plan is limited to 10 sandboxes","Idle sandboxes stop after 10 minutes by default and stopped ones are deleted after 14 days","Billing is in LSU credits shared with other LangSmith products"],"sources":[{"url":"https://www.langchain.com/pricing-langsmith","quote":"Compute · 0.0576 LSU / vCPU-hr","checked":"2026-11-13"},{"url":"https://www.langchain.com/pricing-langsmith","quote":"Memory · 0.01845 LSU / GiB-hr","checked":"2026-11-13"},{"url":"https://www.langchain.com/pricing-langsmith","quote":"Storage · 0.000123 LSU / GiB-hr","checked":"2026-11-13"},{"url":"https://www.langchain.com/pricing-langsmith","quote":"All rates billed per second.","checked":"2026-11-13"},{"url":"https://www.langchain.com/pricing-langsmith","quote":"Includes 8 LSU of Sandbox usage / mo for free.","checked":"2026-11-13"},{"url":"https://www.langchain.com/pricing-langsmith","quote":"Limited to 10 Sandboxes.","checked":"2026-11-13"},{"url":"https://www.langchain.com/pricing-langsmith","quote":"1 LSU = $1.","checked":"2026-11-13"},{"url":"https://www.langchain.com/langsmith/sandboxes","quote":"Each one uses microVM isolation and is ephemeral by default.","checked":"2026-11-13"},{"url":"https://docs.langchain.com/langsmith/sandbox-sdk","quote":"| CPU | 0.5 vCPU | 0.05 to 16 vCPU.","checked":"2026-11-13"},{"url":"https://docs.langchain.com/langsmith/sandbox-sdk","quote":"| Memory | 4 GiB per vCPU | Up to 64 GiB.","checked":"2026-11-13"},{"url":"https://docs.langchain.com/langsmith/sandbox-sdk","quote":"a stopped sandbox wakes on the next command, file operation, or service-URL request.","checked":"2026-11-13"},{"url":"https://docs.langchain.com/langsmith/sandboxes","quote":"Restricted egress applies to non-Enterprise organizations by default and to organizations that request it.","checked":"2026-11-13"},{"url":"https://docs.langchain.com/langsmith/sandboxes","quote":"Sandboxes are generally available in all LangSmith Cloud regions","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/langsmith-sandboxes","score":{"code-sandbox":{"total":72,"rank":13,"of":63}},"comparablePrice":{"code-sandbox":33.263999999999996},"benchmarks":[]},{"slug":"lightpanda","name":"Lightpanda Cloud","company":"Lightpanda","website":"https://lightpanda.io","docsUrl":"https://lightpanda.io/docs","pricingUrl":"https://lightpanda.io/pricing","categories":["browser"],"tagline":"Hosted Lightpanda, a headless browser written from scratch in Zig, with 300 hours for $19 then $0.08 per hour.","summary":"Lightpanda is a headless browser engine built from scratch in Zig with V8 for JavaScript, made for automation, scraping and agents rather than for display. Lightpanda Cloud runs it for you behind a CDP endpoint that Puppeteer, Playwright or chromedp connect to, plus an HTTP API and an MCP server, in west Europe and west US. The free Explorer plan includes 10 browser hours a month; Builder is $19 a month for 300 hours, then $0.08 per hour. The browser itself is AGPL-3.0 open source and can be run locally.","isolation":"Not stated","isolationKind":"unknown","openSource":true,"selfHost":true,"license":"AGPL-3.0","repoUrl":"https://github.com/lightpanda-io/browser","lastRelease":"2026-11-13","sdks":["CDP (Puppeteer, Playwright, chromedp)","HTTP API","MCP server"],"regions":"West Europe (euwest) and west US (uswest).","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not stated","billedWhile":"Browser hours; how session time is measured is not stated","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.08},"gpu":[],"plans":[{"name":"Explorer","monthlyFee":0,"includedUsageUsd":0,"limits":"10 browser hours per month; 5 concurrent sessions; bring your own proxy; community support on Discord. Overage pricing is not stated.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Builder","monthlyFee":19,"includedUsageUsd":0,"limits":"300 browser hours per month, then $0.08 per hour; 30 concurrent sessions; bring your own proxy; email and Slack support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Explorer plan: 10 browser hours every month, free with no end date, 5 concurrent sessions.","notes":["The per-hour price shown is the Builder overage rate of $0.08 after the 300 included hours. Using all included hours, Builder works out to about $0.063 per hour.","Both self-serve plans use your own proxy; the cloud also offers shared proxies, but their price is not stated on the pricing page.","The cloud runs the Lightpanda engine by default. Google Chrome can be selected for debugging but is not meant for production traffic, and whether Chrome hours are priced differently is not stated.","Lightpanda does not render pages visually, so it is not a fit for screenshot-based computer-use agents.","Enterprise (custom volume pricing, SLA, on-premises or private cloud) is priced on request.","The browser engine is AGPL-3.0 and free to run yourself."]},"bestFor":["High-volume scraping and agent page reads over CDP where a lighter, faster browser helps","Teams that bring their own proxies","Running the same engine locally for free and in the cloud for scale"],"watchOut":["It is not Chromium, so some sites or scripts may behave differently","No visual rendering, so screenshots and pixel-based agents are not supported","Only two regions","Billing increment and session time limits are not published"],"sources":[{"url":"https://lightpanda.io/pricing","quote":"300 hours included then +$0.08/hr","checked":"2026-11-13"},{"url":"https://lightpanda.io/pricing","quote":"10 hours included monthly","checked":"2026-11-13"},{"url":"https://lightpanda.io/pricing","quote":"Bring your own proxy","checked":"2026-11-13"},{"url":"https://lightpanda.io/docs/cloud-offer/getting-started","quote":"wss://uswest.cloud.lightpanda.io/ws (west US)","checked":"2026-11-13"},{"url":"https://lightpanda.io/docs/cloud-offer/getting-started","quote":"For debugging, append browser=chrome to run Google Chrome instead, since Chrome isn’t intended for production traffic.","checked":"2026-11-13"},{"url":"https://lightpanda.io/llms.txt","quote":"Lightpanda is a headless browser engine written in Zig, purpose-built for web automation, web scraping, and AI agents.","checked":"2026-11-13"},{"url":"https://github.com/lightpanda-io/browser","quote":"AGPL-3.0 license","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/lightpanda","score":{"browser":{"total":43,"rank":8,"of":12}},"comparablePrice":{"browser":0.08},"benchmarks":[]},{"slug":"lizard","name":"Lizard Sandboxes","company":"Lizard","website":"https://lizard.build","docsUrl":"https://lizard.build/docs/sandboxes/","pricingUrl":"https://lizard.build/pricing","categories":["code-sandbox"],"tagline":"Firecracker microVM sandboxes at flat per-size hourly prices from $0.0081 for 1 vCPU / 2 GB, billed per second.","summary":"Lizard Sandboxes are Firecracker microVMs with their own kernel, created from TypeScript or Python SDKs, the lizard CLI, the dashboard or an HTTP API in US East or EU West. Each size has a flat hourly price that covers CPU, RAM and local disk, from $0.00405 for 0.5 vCPU / 1 GB to $0.1296 for 16 vCPU / 32 GB, billed per second including idle time. Pausing stops compute charges and keeps the files, but not memory or running processes; a paused sandbox costs $0.14 per GB-month of used disk. Running sandboxes can be forked with memory included. Usage comes from a Pro plan that costs $19 a month and includes $19 of usage.","isolation":"Firecracker microVM per sandbox with its own kernel","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","Python","CLI","REST API"],"regions":"US East (Virginia, us-east-1) and EU West (Limburg), chosen per sandbox. The trial runs only in US regions.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":false,"networkControls":null,"maxSessionHours":596,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.0081,"instance":"Mini: 1 vCPU / 2 GB RAM / 10 GB disk","storageGibMonth":0.14,"billingIncrement":"per second","billedWhile":"Wall-clock while the sandbox runs, idle time included. Pausing stops compute charges; a paused sandbox and each snapshot cost $0.14 per GB-month of used disk. Persistent Volumes bill separately.","cpuActiveOnly":false,"unitVcpu":1,"unitGib":2,"browserHour":null},"gpu":[],"plans":[{"name":"Pro","monthlyFee":19,"includedUsageUsd":19,"limits":"$19 of usage a month across apps, databases and sandboxes, then pay as you go. Up to 40 vCPU of sandboxes running at once, unlimited starts.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro 5x","monthlyFee":99,"includedUsageUsd":99,"limits":"$99 of usage a month. Up to 220 vCPU of sandboxes at once, unlimited starts.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro 20x","monthlyFee":499,"includedUsageUsd":499,"limits":"$499 of usage a month. Up to 1,200 vCPU of sandboxes at once, unlimited starts.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"7-day Pro trial with $5 of usage; a card is required but nothing is charged during the 7 days. The trial runs at most 1 Small, 2 Mini and 2 Nano sandboxes at a time, in US regions only, and deletes each after at most 1 hour.","notes":["Per-size prices are linear at $0.0081 per hour for each 1 vCPU / 2 GB: Nano 0.5/1 GB $0.00405, Mini 1/2 GB $0.0081, Small 2/4 GB $0.0162, Medium 4/8 GB $0.0324 (default), Large 8/16 GB $0.0648, Big 16/32 GB $0.1296 (paid plans only).","Lizard states memory in GB without saying GB or GiB; we use the GB figures as GiB.","Local disk per size: Nano 5 GB, Mini 10 GB, Small 20 GB, Medium 40 GB, Large 80 GB, Big 120 GB.","Persistent Volumes, paused sandboxes and snapshots cost $0.14 per GB-month of used disk (the storage rate we show). Sandbox egress is included up to 2 TB per account over a rolling 30 days.","Limits count vCPU of running sandboxes across all of an owner's workspaces; paused sandboxes do not count. Medium, Large and Big unlock after the first paid invoice. The trial allows 50 Small, 100 Mini and 100 Nano starts in total.","Lifetime is set per sandbox (5 minutes by default from the SDK and CLI, none from the raw API); with no deadline a VM stops after 596 hours.","Pause keeps the files only: memory and running processes end, and a resume boots the sandbox fresh. Fork copies a running sandbox including memory. Templates come from a fixed catalog with no custom-template upload. Outbound network controls are not described.","No GPUs and no self-hosting."]},"bestFor":["Large fleets of small sandboxes where a low flat hourly price matters more than idle-time billing","Agents that fork a running sandbox with memory and processes included","Teams that also want to host apps and databases on the same plan"],"watchOut":["Running sandboxes bill for idle time; pausing stops compute but keeps billing for used disk, and memory is not kept","Pro caps sandboxes at 40 vCPU running at once","No custom image upload; templates come from Lizard's catalog","The trial needs a card and limits sandboxes to small sizes in the US"],"sources":[{"url":"https://lizard.build/pricing","quote":"Mini (1 vCPU, 2 GB RAM) at $0.0081/hour","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"1 vCPU · 2 GB RAM · 10 GB disk","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"Large 8 vCPU · 16 GB RAM · 80 GB disk |$0.0648 per hour","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"Pro costs $19 a month and includes $19 of usage for the whole platform: apps, databases, Sandboxes, storage and egress.","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"Pro 5x ($99) and Pro 20x ($499) work the same way, include $99 and $499 of usage, and raise the sandbox limits.","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"Pro starts with a 7-day trial that includes $5 of usage. You need a card to start it, but we charge nothing during the 7 days","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"Persistent Volumes |$0.14 per GB/month","checked":"2026-11-15"},{"url":"https://lizard.build/docs/platform/limits/","quote":"Pro runs up to 40 vCPU of sandboxes at once, Pro 5x 220 and Pro 20x 1,200.","checked":"2026-11-15"},{"url":"https://lizard.build/docs/platform/limits/","quote":"Pro, Pro 5x and Pro 20x have no limit on sandbox starts.","checked":"2026-11-15"},{"url":"https://lizard.build/docs/platform/limits/","quote":"The trial runs 1 Small, 2 Mini and 2 Nano sandboxes at a time, only in US regions, and deletes each sandbox 1 hour after it starts at the latest.","checked":"2026-11-15"},{"url":"https://lizard.build/docs/sandboxes/","quote":"Sandboxes run as Firecracker microVMs, each with its own kernel.","checked":"2026-11-15"},{"url":"https://lizard.build/docs/sandboxes/","quote":"The public CLI does not provide a custom-template upload flow.","checked":"2026-11-15"},{"url":"https://lizard.build","quote":"For longer tasks you can remove the deadline; the VM then stops after 596 hours (about 24 days).","checked":"2026-11-15"},{"url":"https://lizard.build","quote":"Sandboxes run on CPUs only.","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"Big (16 vCPU, 32 GB RAM; paid plans only) at $0.1296/hour","checked":"2026-11-15"},{"url":"https://lizard.build/pricing","quote":"A paused sandbox costs $0.14 per GB-month of used disk space until you resume or delete it, and so does each snapshot until you delete it.","checked":"2026-11-15"},{"url":"https://lizard.build/docs/sandboxes/","quote":"Running processes and memory end, and a resume boots the sandbox fresh with its files.","checked":"2026-11-15"},{"url":"https://lizard.build/docs/sandboxes/","quote":"fork copies a running sandbox, memory and processes included","checked":"2026-11-15"},{"url":"https://lizard.build/docs/platform/limits/","quote":"During the trial, you can start 50 Small, 100 Mini and 100 Nano sandboxes in total across all your workspaces.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/lizard","score":{"code-sandbox":{"total":76,"rank":4,"of":63}},"comparablePrice":{"code-sandbox":19},"benchmarks":[]},{"slug":"maritime","name":"Maritime","company":"Maritime","website":"https://maritime.sh","docsUrl":"https://maritime.sh/docs","pricingUrl":"https://maritime.sh/pricing","categories":["dev-box","code-sandbox"],"tagline":"Per-second micro-VM sandboxes plus persistent agent boxes at a flat monthly price per 2 GB of memory.","summary":"Maritime offers two kinds of headless Linux micro-VM for agents. Sandboxes are short-lived machines billed per second at $0.0055 per vCPU-hour plus $0.00275 per GB-hour while running, ended with their disk after an idle period or a time limit of up to 24 hours. Agent boxes are persistent Firecracker micro-VMs that keep /data, sleep when idle and wake in about a second, for a flat $1 a month per 2 GB of memory with unlimited awake hours. Every plan fee becomes the same amount of usage credit, and usage beyond it is billed at the same rates.","isolation":"Micro-VM per machine with its own kernel and disk (Firecracker for agent boxes)","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","CLI","REST API","MCP","E2B-compatible API"],"regions":"Not stated on the pages checked.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.0055,"gibHour":0.00275,"instanceHour":null,"instance":null,"storageGibMonth":0.08,"billingIncrement":"per second, no minimum","billedWhile":"Sandboxes: wall-clock while running; a paused sandbox pays only for its disk. Agent boxes: a flat monthly price per 2 GB of memory, awake or asleep.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Starter","monthlyFee":20,"includedUsageUsd":20,"limits":"$20 of usage credit a month; 5 machines awake at once. Usage beyond the credit is billed at the same rates.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Growth","monthlyFee":100,"includedUsageUsd":100,"limits":"$100 of usage credit a month; 25 machines awake at once; per-end-user cost reporting. Usage beyond the credit is billed at the same rates.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":500,"includedUsageUsd":500,"limits":"$500 of usage credit a month; 60 machines awake at once; Swarms. Usage beyond the credit is billed at the same rates.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Dedicated","monthlyFee":5000,"includedUsageUsd":5000,"limits":"$5,000 of usage credit a month; 500 machines awake at once; swarm runs up to 10,000 sandboxes; setup of apps, images and network by the vendor; BAA and DPA.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"7-day free trial on Starter terms with up to $20 of usage. A card is needed to start the first machine but is not charged unless the trial is not cancelled, in which case $20 is charged on day 7 as that month's credit.","notes":["Rates shown are sandbox rates: $0.0055 per vCPU-hour + $0.00275 per GB-hour, the same on every plan (default small size, 2 vCPU / 4 GB, is $0.022 per hour). Memory is stated in GB; we treat it as GiB.","Agent boxes are priced differently: $1 a month per 2 GB of memory (1 vCPU at 2 GB; 4 GB is $2, 8 GB is $4), billed per second over a 730-hour month whether awake or asleep, with unlimited awake hours and 5 GB of disk included. That works out to about $0.00137 per hour per 2 GB box. Self-serve agent boxes go up to 8 GB RAM and 100 GB disk.","Storage at $0.08 per GB-month applies to agent box disk beyond 5 GB and to the disk of a sleeping or paused machine. Sandbox disk is included while running.","Sandboxes end, with their disk, after 1 to 60 idle minutes (default 5) or at a time limit of 5 minutes to 24 hours (default 24 hours), unless set to pause when idle. Agent boxes have no time limit and are deleted only by the user. Sandbox sizes go up to 16 vCPU and 64 GB.","Each plan fee is returned one-for-one as usage credit that does not roll over. Usage beyond the credit is billed with no cap; plans differ mainly in how many machines can be awake at once.","Swarms (Scale and Dedicated) rent a whole 500 GB swarm server for $1 a minute while rented; sandboxes on it have no separate charge.","Maritime also sells Linux desktops ($0.12/hour), Windows desktops ($0.18/hour, Growth and up) and virtual Android phones ($0.15/hour), and managed agent hosting with chat channels and triggers. Those are not covered here.","AI model credits are bought separately at provider list price; bringing your own key is free."]},"bestFor":["One persistent machine per end user or per agent at a flat low monthly price, with unlimited awake hours.","Short-lived code execution at low per-second rates, including existing E2B SDK code pointed at Maritime.","Mixing sandboxes and persistent boxes under one plan credit and invoice."],"watchOut":["Agent boxes are billed every month whether awake or asleep, and self-serve sizes stop at 8 GB of memory.","Sandboxes end with their disk after a few idle minutes by default; background commands do not count as activity.","The plan limits how many machines can be awake at once (5 on Starter); further starts are refused.","Usage beyond the plan credit has no spending cap."],"sources":[{"url":"https://maritime.sh/pricing","quote":"|Sandbox |$0.0055 per vCPU-hour + $0.00275 per GB-hour Per second, no minimum |","checked":"2026-11-14"},{"url":"https://maritime.sh/pricing","quote":"|Agent box |$1 per 2 GB a month Monthly, awake or asleep, 5 GB of disk included |","checked":"2026-11-14"},{"url":"https://maritime.sh/pricing","quote":"|Disk of a sleeping machine |$0.08 per GB-month Per second while asleep |","checked":"2026-11-14"},{"url":"https://maritime.sh/pricing","quote":"One for one. A $100 Growth plan gives you $100 of usage each month.","checked":"2026-11-14"},{"url":"https://maritime.sh/pricing","quote":"The first 7 days are free, with up to $20 of usage on Starter terms.","checked":"2026-11-14"},{"url":"https://maritime.sh/pricing","quote":"Any size up to 16 vCPU and 64 GB","checked":"2026-11-14"},{"url":"https://maritime.sh/pricing","quote":"Usage past your plan credit is billed at the same rates with no cap","checked":"2026-11-14"},{"url":"https://maritime.sh/docs/ephemeral-sandboxes.md","quote":"Idle time 1 to 60 minutes, time limit 5 to 1,440 minutes.","checked":"2026-11-14"},{"url":"https://maritime.sh/docs/ephemeral-sandboxes.md","quote":"Set `E2B_DOMAIN` to `sb.mtm.sh` and `E2B_API_KEY` to a Maritime key with the `sandboxes` scope.","checked":"2026-11-14"},{"url":"https://maritime.sh/docs/persistent-sandboxes.md","quote":"Each one is its own Firecracker micro-VM with its own disk.","checked":"2026-11-14"},{"url":"https://maritime.sh/docs/persistent-sandboxes.md","quote":"2 GB RAM and 5 GB SSD by default; self-serve up to 8 GB RAM and 100 GB SSD.","checked":"2026-11-14"},{"url":"https://maritime.sh","quote":"Private networks, internet policies and warm pools","checked":"2026-11-14"},{"url":"https://maritime.sh/docs/ephemeral-sandboxes.md","quote":"a stock template such as `base` or `python`, a saved snapshot, or an OCI image","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/maritime","score":{"dev-box":{"total":61,"rank":22,"of":38},"code-sandbox":{"total":59,"rank":41,"of":63}},"comparablePrice":{"dev-box":20,"code-sandbox":20},"benchmarks":[]},{"slug":"microsandbox","name":"microsandbox","company":"Super Rad Company","website":"https://github.com/superradcompany/microsandbox","docsUrl":"https://docs.microsandbox.dev","pricingUrl":"https://github.com/superradcompany/microsandbox","categories":["code-sandbox"],"tagline":"Open-source library and CLI that boots local microVMs from OCI images on Linux, macOS and Windows.","summary":"microsandbox runs untrusted code in small hardware-isolated virtual machines on your own machine, started from the msb CLI or embedded directly in an application through its SDKs, with no server or daemon to run. Sandboxes boot from standard OCI container images, can be forked while running, snapshotted and restored, and can be limited to allowed hosts with secrets that never enter the VM. It is free Apache-2.0 software and the project describes itself as beta.","isolation":"Hardware-isolated microVM per sandbox (KVM on Linux, Apple Silicon virtualization on macOS, Windows Hypervisor Platform on Windows)","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/superradcompany/microsandbox","lastRelease":"2026-11-12","sdks":["TypeScript","Python","Rust","Go","Ruby","CLI","MCP server"],"regions":"Runs on your own hardware: Linux with KVM, macOS on Apple Silicon, or Windows with Windows Hypervisor Platform.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own hardware","billedWhile":"Not applicable: you pay only for your own hardware or cloud machines","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license; no hosted paid tier with a published price was found.","notes":["No usage charges: the software is free and runs on hardware you provide.","The project states it is beta software and warns of breaking changes.","Sandboxes can be stopped and started, forked while running, and saved to a snapshot file that is restored later."]},"bestFor":["Running agent or user code in microVMs on a laptop or single server without a separate daemon","Embedding sandbox creation directly in TypeScript, Python, Rust, Go or Ruby applications","Workflows that fork or snapshot a running sandbox and restore it later"],"watchOut":["Beta software: breaking changes and missing features are expected","Needs hardware virtualization: KVM on Linux, Apple Silicon on macOS, or Windows Hypervisor Platform","You run and scale the hosts yourself; there is no multi-node scheduler described"],"sources":[{"url":"https://github.com/superradcompany/microsandbox","quote":"Microsandbox runs untrusted workloads inside fast, local microVMs","checked":"2026-11-13"},{"url":"https://github.com/superradcompany/microsandbox","quote":"Fork live sandboxes","checked":"2026-11-13"},{"url":"https://github.com/superradcompany/microsandbox","quote":"Save running sandbox state and restore later.","checked":"2026-11-13"},{"url":"https://github.com/superradcompany/microsandbox","quote":"Runs standard container images from Docker Hub, GHCR, or any OCI registry.","checked":"2026-11-13"},{"url":"https://github.com/superradcompany/microsandbox","quote":"Spawn VMs right within your code. No setup server. No long-running daemon.","checked":"2026-11-13"},{"url":"https://github.com/superradcompany/microsandbox","quote":"Expect breaking changes, missing features, and rough edges.","checked":"2026-11-13"},{"url":"https://github.com/superradcompany/microsandbox/releases/tag/v0.7.15","quote":"v0.7.15","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/microsandbox","score":{"code-sandbox":{"total":68,"rank":24,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"microsoft-nvx","name":"Microsoft NVX","company":"Microsoft Research","website":"https://github.com/microsoft/nvx","docsUrl":"https://github.com/microsoft/nvx","pricingUrl":"https://github.com/microsoft/nvx","categories":["code-sandbox"],"tagline":"MIT-licensed micro-VM sandbox from Microsoft Research, built on OpenVMM, for KVM, MSHV and Windows hosts.","summary":"NVX is an ultra-light micro-VM sandbox for running untrusted workloads with hardware isolation, built on OpenVMM with a Linux guest. It was developed by Microsoft Research groups and runs on Linux with KVM or MSHV and on Windows with the Windows Hypervisor Platform. The nvx-rs command-line tool and Rust crate launch and supervise microVMs, with Alpine, Ubuntu and Azure Linux userlands included. Only development releases have been published so far.","isolation":"Hardware-isolated micro-VM on OpenVMM with a Linux guest (KVM, MSHV or Windows Hypervisor Platform)","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"MIT","repoUrl":"https://github.com/microsoft/nvx","lastRelease":"2026-11-13","sdks":["Rust crate","CLI (nvx-rs)"],"regions":"Runs on your own Linux (KVM or MSHV) or Windows (WHP) host.","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own hardware","billedWhile":"Not applicable: no usage charges","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the MIT license.","notes":["All published GitHub releases are development prereleases (0.1.0-dev builds); the date shown is the most recent one.","The README covers launching single microVMs; it does not describe a multi-host service, snapshots or network policy features."]},"bestFor":["Building your own sandbox service on a small micro-VM that runs on both Linux and Windows hosts","Research and experiments with OpenVMM-based isolation","Rust projects that want to launch and supervise microVMs from code"],"watchOut":["Development releases only; no stable version yet","A building block: no hosted service, scheduler or SDKs beyond Rust","Needs KVM, MSHV or Windows Hypervisor Platform on the host"],"sources":[{"url":"https://github.com/microsoft/nvx","quote":"NVX is an ultra-light micro-VM sandbox for running untrusted workloads with hardware-enforced isolation. It is built on top of OpenVMM and runs Linux as a guest.","checked":"2026-11-15"},{"url":"https://github.com/microsoft/nvx","quote":"The package also includes the Ubuntu and Azure Linux userlands","checked":"2026-11-15"},{"url":"https://github.com/microsoft/nvx","quote":"Requires Windows Hypervisor Platform enabled.","checked":"2026-11-15"},{"url":"https://github.com/microsoft/nvx/releases","quote":"v0.1.0-dev","checked":"2026-11-15"},{"url":"https://github.com/microsoft/nvx/blob/dev/LICENSE","quote":"MIT License","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/microsoft-nvx","score":{"code-sandbox":{"total":57,"rank":46,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"miosa","name":"MIOSA","company":"MIOSA","website":"https://miosa.ai","docsUrl":"https://miosa.ai/docs/","pricingUrl":"https://miosa.ai/pricing","categories":["dev-box","code-sandbox","browser"],"tagline":"Firecracker sandboxes, persistent desktop computers and deployments from one API, billed per second by machine size.","summary":"MIOSA offers isolated sandboxes for code, persistent desktop computers for browser and computer-use agents, and a path to publish the result as a deployment with databases and custom domains. Sandboxes run in Firecracker microVMs by default, keep their disk when stopped, and can pause, snapshot and fork. Sandboxes and desktops are billed a flat price per size, per second while running, from $0.0316 an hour for 1 vCPU / 0.5 GB to $0.7141 for 16 vCPU / 32 GB. Every plan, including the free one, credits its monthly price back as usage.","isolation":"Firecracker microVM by default; an alternative SOMA runtime is also offered","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","Go","Rust","Java","Elixir","CLI","MCP server","REST API"],"regions":"US West (Los Angeles), US East (New York) and US East 2 (Miami). Bring-your-own-cloud on AWS or Google Cloud is set up with MIOSA.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.0893,"instance":"small: 2 vCPU, 4 GB","storageGibMonth":0.09,"billingIncrement":"Per second","billedWhile":"Flat price of the size while the sandbox or desktop runs; stopped or paused machines pay no compute, only disk.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":4,"browserHour":0.0446},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":1,"limits":"$1 of usage every month. 5 sandboxes and 1 computer at once, machines up to 4 vCPU / 8 GB, 8 GB and 4 vCPU running in total, 5 GB egress a month, 1 seat. Machines pause at a $0 balance; credits can be added from $5.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Developer","monthlyFee":20,"includedUsageUsd":20,"limits":"$20 a month counted toward usage. 50 sandboxes and 5 computers at once, machines up to 4 vCPU / 8 GB, 128 GB and 64 vCPU running in total, 100 GB egress a month, 1 seat.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Business","monthlyFee":500,"includedUsageUsd":500,"limits":"$500 a month counted toward usage. 250 sandboxes and 50 computers at once, machines up to 8 vCPU / 16 GB, 512 GB and 256 vCPU running in total, 1 TB egress a month, 10 seats, dedicated vCPU sizes.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Enterprise","monthlyFee":2000,"includedUsageUsd":2000,"limits":"$2,000 a month counted toward usage. 1,000 sandboxes and 200 computers at once, machines up to 16 vCPU / 32 GB, 1,024 GB and 512 vCPU running in total, 5 TB egress a month, unlimited seats.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: $1 of usage every month with no overage; machines pause at a $0 balance.","notes":["Sandbox sizes and hourly prices: micro 1 vCPU / 0.5 GB $0.0316, tiny 1 / 1 GB $0.0359, xs 1 / 2 GB $0.0446, small 2 / 4 GB $0.0893, quad 4 / 4 GB $0.1437, medium 4 / 8 GB $0.1785, large 8 / 16 GB $0.357, xl 16 / 32 GB $0.7141. The price is not linear in a single unit, so we show the small size.","Desktop computers cost the same as a sandbox of the same size (xs to xl); the browser-hour price we show is the xs desktop at $0.0446.","Storage: disk $0.09 per GB-month while it exists, snapshots $0.018 per GB-month, object storage $0.0162 per GB-month. 10 named snapshots per organization are included; each extra one costs $1.70 a month. Egress is $0.054 per GB from the first GB.","Sandboxes stop after an idle-based timeout (default 3,600 seconds, at most 86,400) unless set to always on; on accounts opened since October 11, 2026 they also stop after 30 idle minutes by default and wake on the next request.","Pause is available only on the Firecracker runtime; the SOMA runtime has no pause or snapshots.","The largest self-serve machine is 16 vCPU / 32 GB, which needs the $2,000 Enterprise plan; Free and Developer stop at 4 vCPU / 8 GB.","MIOSA states memory in GB without saying GB or GiB."]},"bestFor":["Agent products that need code sandboxes, desktops and production deployments under one account","Small, cheap sandboxes for short code runs, from about $0.03 an hour","Computer-use agents that need a desktop priced the same as a sandbox"],"watchOut":["Sizes above 4 vCPU / 8 GB need the $500 or $2,000 monthly plan","Egress is billed from the first GB","Only US regions","No GPU sandboxes yet"],"sources":[{"url":"https://miosa.ai/pricing","quote":"small Popular 2 4 GB $0.0893 /h $65.16 /mo","checked":"2026-11-13"},{"url":"https://miosa.ai/pricing","quote":"A computer costs the same as a sandbox of the same size, billed the same way.","checked":"2026-11-13"},{"url":"https://miosa.ai/pricing","quote":"Storage attached to a machine, while it exists","checked":"2026-11-13"},{"url":"https://miosa.ai/pricing","quote":"Free has no monthly charge and includes $1 of usage every month.","checked":"2026-11-13"},{"url":"https://miosa.ai/pricing","quote":"Sandboxes and computers are billed the flat price of their size, per second, while they run.","checked":"2026-11-13"},{"url":"https://miosa.ai/pricing","quote":"200 computers · machines up to 16 vCPU / 32 GB","checked":"2026-11-13"},{"url":"https://miosa.ai/docs/develop/sandboxes.md","quote":"runtime: firecracker (the default), soma, or auto.","checked":"2026-11-13"},{"url":"https://miosa.ai/docs/regions.md","quote":"All three are generally available and none is restricted.","checked":"2026-11-13"},{"url":"https://miosa.ai/docs/develop/sandboxes/timeouts.md","quote":"It defaults to 3600 seconds and accepts values from 1 through 86400.","checked":"2026-11-13"},{"url":"https://miosa.ai/docs/known-limitations/","quote":"The public sandbox create contract does not accept, persist, or schedule GPU requests","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/miosa","score":{"dev-box":{"total":77,"rank":6,"of":38},"code-sandbox":{"total":75,"rank":8,"of":63},"browser":{"total":81,"rank":1,"of":12}},"comparablePrice":{"dev-box":20,"code-sandbox":20,"browser":0.0446},"benchmarks":[]},{"slug":"modal","name":"Modal Sandboxes","company":"Modal Labs","website":"https://modal.com","docsUrl":"https://modal.com/docs/guide/sandboxes","pricingUrl":"https://modal.com/pricing","categories":["code-sandbox","gpu"],"tagline":"Containers for untrusted code on Modal's serverless platform, billed per second on max(request, actual usage).","summary":"Modal Sandboxes are containers created from the Python, JavaScript or Go SDK that run on either a gVisor runtime or a VM runtime with its own Linux kernel. They are billed per second at Modal's non-preemptible CPU and memory rates (3x the Function list price), charged on whichever is higher of the requested or actually used resources, so an unrequested sandbox pays mostly for what it uses. Sandboxes can attach GPUs (gVisor runtime only), run up to 24 hours, and save state through filesystem, directory and memory snapshots. Plans add a monthly compute credit: $30 on Starter and $100 on the $250/month Team plan.","isolation":"gVisor container runtime, or a VM runtime with its own Linux kernel (selectable per sandbox)","isolationKind":"gvisor","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","Go"],"regions":"Runs in Modal's multi-cloud fleet; optional region pinning (us, eu, ap and narrower regions such as us-west, uk, jp) at a 1.15x (broad) or 1.75x (narrow) price multiplier.","features":{"pauseResume":false,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.070956,"gibHour":0.024012,"instanceHour":null,"instance":null,"storageGibMonth":0.09,"billingIncrement":"per second","billedWhile":"While the sandbox runs, charged on max(requested, actually used) CPU and memory; default request is 0.125 physical cores and 128 MiB.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"Nvidia B300","perHour":7.0992},{"type":"Nvidia B200","perHour":6.2496},{"type":"Nvidia H200 SXM","perHour":4.5396},{"type":"Nvidia H100 SXM5","perHour":3.9492},{"type":"Nvidia RTX PRO 6000","perHour":3.0312},{"type":"Nvidia A100 80 GB","perHour":2.4984},{"type":"Nvidia A100 40 GB","perHour":2.0988},{"type":"Nvidia L40S","perHour":1.9512},{"type":"Nvidia A10","perHour":1.1016},{"type":"Nvidia L4","perHour":0.7992},{"type":"Nvidia T4","perHour":0.5904}],"plans":[{"name":"Starter","monthlyFee":0,"includedUsageUsd":30,"limits":"$30/month compute credit, up to 3 seats, 100 containers and 10 GPU concurrency, 1 TiB/month free egress, 1-day log retention.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":250,"includedUsageUsd":100,"limits":"$100/month compute credit, unlimited seats, 5,000 containers and 50 GPU concurrency, 10 TiB/month free egress, custom domains for sandbox tunnels, static IP proxy.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Starter plan: $0/month with $30 of compute credit every month (CPU, memory, GPU for Functions and Sandboxes).","notes":["Sandbox rates differ from Function rates: sandboxes are billed at non-preemptible rates of 3x list price. Sandbox CPU is $0.00003942 per physical core per second ($0.141912/core-hour) and memory $0.00000667 per GiB-second ($0.024012/GiB-hour). Function list rates are $0.0000131/core/s ($0.04716/core-hour) and $0.00000222/GiB/s ($0.007992/GiB-hour).","Modal prices CPU per physical core, and one core equals 2 vCPU. Our vCPU-hour figure is the core rate divided by 2 ($0.141912 / 2 = $0.070956).","Billing is max(request, actual) for CPU and memory, with a minimum of 0.125 cores and 128 MiB per container. With no request set, a sandbox pays roughly for its average usage; a request is billed for the sandbox's whole lifetime even if idle. Disk requests add to the memory request at a 20:1 ratio.","GPU sandboxes are billed at the standard GPU rates listed above (per-second rates x3600), only run on the gVisor runtime, and are subject to preemption.","Pinning a region multiplies the price by 1.15x (broad region such as us) or 1.75x (narrow region such as us-west).","Volumes cost $0.09/GiB-month with up to 1 TiB free; egress costs $0.04/GiB after the plan's included egress (1, 10 or 100 TiB/month).","Sandbox timeout defaults to 5 minutes and can be set up to 24 hours; there is no pause/resume, but filesystem snapshots (30-day default TTL) and memory snapshots (alpha, 7-day expiry, terminates the sandbox) can carry state to a new sandbox.","No maximum CPU or memory per sandbox is published on the pages checked; Modal only says requests above a maximum are rejected."]},"bestFor":["Bursty agent workloads where average CPU and memory use is far below peak, since billing follows actual usage when no request is set.","Sandboxes that need GPUs, from T4 up to B300.","Teams already running Python, JS or Go workloads on Modal who want sandboxes on the same account and credit."],"watchOut":["Sandbox CPU and memory cost 3x Modal's Function list rates.","No pause/resume: sandboxes end after at most 24 hours and state must be carried over with snapshots.","GPU sandboxes can be preempted and do not support memory snapshots.","Region pinning adds a 1.15x to 1.75x price multiplier."],"sources":[{"url":"https://modal.com/pricing","quote":"Sandboxes are billed at non-preemptible rates of 3x list price.","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"Physical core (2 vCPU equivalent) $0.00003942","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"Memory (non-preemptible) $0.00000667","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"Nvidia H100 SXM5 $0.001097","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"$250 + compute / month","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"$30 / month free credits","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"$100 / month free credits","checked":"2026-11-14"},{"url":"https://modal.com/pricing","quote":"Volumes $0.09","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/resources","quote":"For CPU and memory, you'll be charged based on whichever is higher: your request or actual usage.","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/resources","quote":"Disk requests are billed by increasing the memory request at a 20:1 ratio.","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/sandboxes","quote":"You can change this by passing a timeout of up to 24 hours to the Sandbox.create(...) function.","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/sandbox-resources","quote":"By default, each Sandbox requests 0.125 CPU cores and 128 MiB of memory","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/sandbox-resources","quote":"Unlike CPU Sandboxes, GPU Sandboxes are subject to preemption.","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/region-selection","quote":"A multiplier on top of our base usage pricing will be applied to any Function or Sandbox that has a container region defined.","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/sandbox-networking","quote":"Modal provides three levels of outbound network restriction","checked":"2026-11-14"},{"url":"https://modal.com/docs/guide/sandbox-snapshots","quote":"Memory Snapshots expire 7 days after creation and cannot currently be extended.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/modal","score":{"code-sandbox":{"total":57,"rank":47,"of":63},"gpu":{"total":58,"rank":5,"of":7}},"comparablePrice":{"code-sandbox":41.88096,"gpu":3.9492},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":451,"unit":"ms","better":"lower","date":"2026-11-13","rank":10,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":48147,"unit":"ms","better":"lower","date":"2026-11-13","rank":4,"of":25},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":15.54,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":8,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":14.8,"unit":"s","better":"lower","date":"2026-09-29","rank":5,"of":11}]},{"slug":"morph","name":"Morph Cloud","company":"Morph Labs (not related to Morph LLM / morphllm.com, which is AutoInfra, Inc.)","website":"https://cloud.morph.so","docsUrl":"https://cloud.morph.so/docs/documentation/overview","pricingUrl":"https://cloud.morph.so/web/pricing","categories":["dev-box","code-sandbox"],"tagline":"VMs that can be snapshotted, paused and branched into many copies, billed in prepaid compute units.","summary":"Morph Cloud runs VMs (MorphVM instances and Devboxes) on its Infinibranch runtime, which snapshots a running VM and starts new instances from it, including many parallel branches. Usage is billed in Morph Compute Units (MCUs) at $0.05 each; a running VM uses the largest of its vCPU count, RAM/4 GiB and disk/16 GiB in MCUs per hour. Paused instances stop using compute and pay only for the snapshot. Paid tiers are monthly MCU bundles at a discount.","isolation":"Virtual machine (MorphVM on Infinibranch; hypervisor not named)","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI","REST API"],"regions":"Not published","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.05,"instance":"1 MCU: 1 vCPU + 4 GB RAM + 16 GB disk; a VM costs max(vCPUs, ceil(RAM/4 GiB), ceil(disk/16 GiB)) MCUs per hour","storageGibMonth":0.0073,"billingIncrement":"per second (rates published per MCU-hour)","billedWhile":"wall-clock while the instance is running; paused instances bill only snapshot storage","cpuActiveOnly":false,"unitVcpu":1,"unitGib":4,"browserHour":null},"gpu":[],"plans":[{"name":"Developer (Free)","monthlyFee":0,"includedUsageUsd":0,"limits":"0 starting MCUs on the pricing page (buy credits to use compute); MorphVM quota up to 64 vCPU, 256 GB RAM, 1024 GB storage; Devboxes: 8 concurrent, 32 total","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":40,"includedUsageUsd":50,"limits":"1,000 MCUs ($50 at the $0.05 standard rate, i.e. $0.04 per MCU); MorphVM quota up to 256 vCPU, 1024 GB RAM, 4096 GB storage; Devboxes: 32 concurrent, 128 total","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":250,"includedUsageUsd":375,"limits":"7,500 MCUs ($375 at the $0.05 standard rate, about $0.0333 per MCU); MorphVM quota up to 1024 vCPU, 4096 GB RAM, 16384 GB storage; Devboxes: 128 concurrent, 512 total","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Conflicting: the pricing page lists the free Developer tier with 0 starting MCUs ('Purchase credits to use compute'), while the Plans doc says 300 MCUs free and the FAQ/Signing Up docs say 300 free compute hours per month. Access may require an invite.","notes":["MCU conversion: 1 MCU = $0.05 standard rate = 1 vCPU-hour + 4 GB RAM-hours + 16 GB disk-hours (bundled, not additive), or 5 TB snapshot-hours. Because a VM is charged the maximum of its three dimensions, per-vCPU and per-GiB rates are not separable; we price by MCU: max(vCPU, ceil(GiB / 4)) MCUs per hour.","Snapshot storage: 5 TB snapshot-hours per MCU = $0.05 / 5,000 GB-hours = $0.00001 per GB-hour = $0.0073 per GB-month at 730 hours. This matches the vendor example: an 8 vCPU / 8 GB RAM / 8 GB disk snapshot costs $0.12/month (16 GB snapshot).","Paid tiers are discounted MCU bundles: Team $40 for 1,000 MCUs (20% discount), Scale $250 for 7,500 MCUs (33% discount). The pricing page calls this a 'starting credit allocation' next to a monthly price; we read it as a monthly allocation, but whether unused MCUs roll over is not stated. Extra usage is billed pay-as-you-go.","The Terms of Service list per-second compute, per-operation snapshot/instance charges, snapshot/lineage storage (GB-hour) and egress as billable; no per-operation or egress rates are published.","Pricing page quotes RAM in GB but the MCU formula divides by GiB; we treat them as the same unit.","GitHub Actions runners bill from the same MCU pool: Small 3 MCU/hr (2 vCPU/4 GB/40 GB), Medium 5 MCU/hr, Large 10 MCU/hr.","Morph LLM (morphllm.com, fast-apply models) is a different company: its site is copyright AutoInfra, Inc., while Morph Cloud docs are copyright Morph Labs."]},"bestFor":["Branching one running VM into many identical copies for parallel agent rollouts or RL environments","Pausing a VM with memory and processes intact and paying only snapshot storage while paused","Teams that can prepay a monthly MCU bundle for a 20-33% lower unit rate"],"watchOut":["Billing uses the largest of vCPU, RAM/4 GiB and disk/16 GiB, so a VM with a big disk or RAM pays for it even with 1 vCPU","Free tier is unclear: pricing page shows 0 MCUs while docs promise 300 free MCUs or compute hours","Egress and per-operation snapshot/instance charges are named in the terms but no rates are published","No published regions, network egress controls, or per-VM size limits"],"sources":[{"url":"https://cloud.morph.so/web/pricing","quote":"MCUs = max(vCPUs, ceil(RAM/4GiB), ceil(disk/16GiB))","checked":"2026-11-14"},{"url":"https://cloud.morph.so/web/pricing","quote":"Based on standard MCU rate of $0.05","checked":"2026-11-14"},{"url":"https://cloud.morph.so/docs/documentation/setup/plans","quote":"1 vCPU-hour + 4 GB RAM-hours + 16 GB disk-hours","checked":"2026-11-14"},{"url":"https://cloud.morph.so/docs/documentation/setup/signing-up","quote":"our $0/month Developer Plan provides 300 free compute hours per month","checked":"2026-11-14"},{"url":"https://cloud.morph.so/docs/documentation/instances/pause-resume","quote":"The VM stops consuming MCUs","checked":"2026-11-14"},{"url":"https://cloud.morph.so/web/legal/terms-of-service","quote":"Per-second compute; per-operation Snapshot / Instance charges; snapshot / lineage storage (GB-hour); egress.","checked":"2026-11-14"},{"url":"https://cloud.morph.so/docs/faq","quote":"Python SDK (requires Python 3.10+)","checked":"2026-11-14"},{"url":"https://cloud.morph.so/docs/blog/developers","quote":"Fully OCI Compatible, run any docker container","checked":"2026-11-14"},{"url":"https://www.morphllm.com/about","quote":"© 2026 AutoInfra, Inc. All rights reserved.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/morph","score":{"dev-box":{"total":66,"rank":16,"of":38},"code-sandbox":{"total":67,"rank":26,"of":63}},"comparablePrice":{"dev-box":22,"code-sandbox":17.6},"benchmarks":[]},{"slug":"mosaic-sandbox","name":"Mosaic Sandbox","company":"Mosaic","website":"https://sandbox.mosaicos.com","docsUrl":"https://sandbox.mosaicos.com/docs/","pricingUrl":"https://sandbox.mosaicos.com/pricing/","categories":["code-sandbox"],"tagline":"Firecracker microVMs at a flat $0.05 per active hour for 2 vCPU / 4 GB, free while hibernated, with native SSH.","summary":"Mosaic Sandbox runs each sandbox as a Firecracker microVM on KVM, driven through a REST API, the mos CLI, Python, TypeScript and Go SDKs, MCP or native SSH. One price covers CPU, RAM and disk: $0.05 per active hour for each 2 vCPU / 4 GB billing unit, with larger shapes billed as multiple units. A quiet sandbox hibernates after three seconds and costs nothing until the next request wakes it, and egress has no per-GB charge. A task mode clones a repository, installs dependencies, runs a command and reports each step.","isolation":"Firecracker microVM on Linux KVM per sandbox; no shared-root containers","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","Go","CLI","MCP server","REST API"],"regions":"Silicon Valley and Northern Virginia, placed automatically with failover or pinned per sandbox. No EU region today.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.05,"instance":"2 vCPU / 4 GB (one billing unit, the default shape)","storageGibMonth":null,"billingIncrement":"per second ($0.00001389 per second per unit)","billedWhile":"Only while the sandbox is awake. It hibernates after 3 quiet seconds and is not billed while hibernated; the next API, file, exec or SSH request wakes it.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":4,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"One SKU for everyone. Up to 1,000 sandboxes alive at once per organization (hibernated ones included). RBAC, SSO and audit logs included.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free tier or trial credit is stated on the pricing or signup pages.","notes":["Billing units are whole: max(ceil(vCPU/2), ceil(MiB/4096)). A 1 vCPU / 2 GB guest still bills as one unit; 4 vCPU / 8 GB bills as two; 16 vCPU / 32 GB as eight.","The $0.05 hourly price includes RAM and the sandbox's storage. Pricing for volumes kept outside a sandbox is not stated.","Shapes: 1/2, 2/4 (default), 4/8, 8/16 and 16/32 vCPU/GB; 8 and 16 vCPU are served on the base template and on custom environments. Only the default shape is kept warm, so larger shapes cold boot.","Mosaic states sizes in GB on the pricing page and MiB in the API; we treat the 4,096 MiB unit as 4 GiB.","Egress has no per-GB charge. Outbound traffic is open by default under a deny-by-default eBPF policy, with allowlists and deny-all per sandbox; SSH cannot be combined with a network-disabled sandbox.","Sandbox lifetime is set per sandbox (examples use up to 24 hours); no maximum is published.","Custom environments can be built from registry images. No GPU support."]},"bestFor":["Agents with long gaps between commands, since hibernation is free and wakes on the next request","Repository tasks that should clone, install and test in one call","Developers who want plain SSH access to a sandbox"],"watchOut":["A 1 vCPU sandbox costs the same as 2 vCPU / 4 GB","Only US regions today","No free tier or credit is stated","No GPU support"],"sources":[{"url":"https://sandbox.mosaicos.com/pricing/","quote":"$0.00001389 / second, at 2 vCPU / 4 GB. The 4 vCPU guest bills as two of those; the 1 vCPU guest still bills as one.","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/pricing/","quote":"Up to 1,000 sandboxes at once per organization.","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/pricing/","quote":"Firecracker microVM · API/CLI/MCP · RBAC · SSO · audit logs · webhooks · OpenTelemetry export · hibernation","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com","quote":"Includes RAM and storage. $0 while hibernated","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/llms.txt","quote":"Billing units are whole: max(ceil(vCPU/2), ceil(MiB/4096)), so 15 vCPU = 8 units and 9216 MiB = 3.","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/llms.txt","quote":"Shapes (self-serve, pick with `memory_mb`/`vcpu`): 1/2048, 2/4096 (default), 4/8192, 8/16384 and 16/32768 vCPU/MiB.","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/llms.txt","quote":"Sandboxes currently run in Silicon Valley and Northern Virginia.","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/llms.txt","quote":"Automatic hibernation begins after three quiet seconds.","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/llms.txt","quote":"It is not the right tool when the workload requires a GPU","checked":"2026-11-13"},{"url":"https://sandbox.mosaicos.com/start/","quote":"Sign in and you have a live `msk_live_…` key, good for the SDKs, the CLI, and the MCP server.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/mosaic-sandbox","score":{"code-sandbox":{"total":74,"rank":10,"of":63}},"comparablePrice":{"code-sandbox":8.8},"benchmarks":[]},{"slug":"namespace","name":"Namespace Devboxes","company":"Namespace Labs","website":"https://namespacelabs.com/devbox","docsUrl":"https://namespace.so/docs/devbox","pricingUrl":"https://namespacelabs.com/pricing","categories":["dev-box","code-sandbox","ci-runner"],"tagline":"Linux, macOS and Windows dev machines for people and coding agents, billed per Devbox Minute plus storage used.","summary":"Namespace Devboxes are remote Linux, macOS and Windows machines on Namespace's own hardware, reachable over SSH, IDEs, a CLI, a TypeScript or Go SDK and MCP. Linux Devboxes run from a container image you choose and include Docker; macOS Devboxes run on Apple Silicon. Persistent Devboxes keep their files across stops and start again when you connect; ephemeral ones are also available. Compute is billed in Devbox Minutes at $0.004 each (a small Linux Devbox uses one per minute) plus $0.20 per GB-month of storage actually used.","isolation":"Namespace compute instance per Devbox on Namespace-operated hardware; the pages checked do not name the virtualization technology","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Go","CLI","MCP server"],"regions":"Runs in Namespace's own sites; the pages checked do not list them (the CLI can measure latency to each Devbox site).","features":{"pauseResume":true,"snapshots":null,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":32,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.24,"instance":"Linux size S: up to 4 vCPU, 8 GB RAM (1 Devbox Minute per minute at $0.004). M, L and XL use 2, 4 and 8 Devbox Minutes per minute.","storageGibMonth":0.2,"billingIncrement":"per minute, in Devbox Minutes","billedWhile":"Compute for each minute the Devbox runs; persistent storage for data actually stored, including while stopped.","cpuActiveOnly":false,"unitVcpu":4,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Developer","monthlyFee":0,"includedUsageUsd":0,"limits":"Pay as you go. 1 Devbox per user, 3 per workspace (running and stopped both count). Max Devbox duration 3 hours.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Devbox Plus","monthlyFee":250,"includedUsageUsd":200,"limits":"$250/month including 50,000 Devbox Minutes and 400 GB-month of storage. 5 Devboxes per user, 50 per workspace. Max Devbox duration 3 hours.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":100,"includedUsageUsd":0,"limits":"$100/month; its included unit minutes cover CI and remote builds, not Devboxes, which are billed separately. 3 Devboxes per user, 30 per workspace. Max Devbox duration 5 hours.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Business","monthlyFee":250,"includedUsageUsd":0,"limits":"$250/month; its included unit minutes cover CI and remote builds, not Devboxes, which are billed separately. 5 Devboxes per user, 50 per workspace. Max Devbox duration 24 hours.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"30-day free trial of the Developer plan on sign-up, including Devboxes; no credit card required. The pricing page does not state how much usage the trial covers.","notes":["Rate shown is for Linux Devboxes: $0.004 per Devbox Minute, and size S (up to 4 vCPU, 8 GB) uses 1 Devbox Minute per minute, so $0.24 an hour. M (8 vCPU / 16 GB) is $0.48, L (16 vCPU / 32 GB) $0.96 and XL (32 vCPU / 64 GB) $1.92 an hour. The vCPU counts are what each size can boost to.","macOS Devboxes come in M (6 vCPU / 14 GB, 15 Devbox Minutes per minute, $3.60 an hour) and L (12 vCPU / 28 GB, 30 per minute, $7.20 an hour). Windows Devboxes are in early access and ephemeral only.","Devbox Plus includes 50,000 Devbox Minutes, which is $200 of compute at list price; the included value shown is that conversion, not a dollar credit. Its 400 GB-month of storage (worth $80) is not counted.","Devbox usage is not drawn from the unit minutes included in Team and Business; those cover CI runners and remote builds, which have their own per-shape prices. CI pricing does not apply to Devboxes.","The pricing page lists Devbox compute as 'Starting at $0.004/min' on Developer, Team and Business. It does not say whether the 1-minute minimum and rounding rules described for CI unit minutes also apply to Devbox Minutes.","Storage is billed on data stored, not on the provisioned volume size. Prices are quoted in GB without saying GB or GiB; we treat them as GiB.","Each Devbox has a maximum duration per run that depends on the plan: 3 hours on Developer and Devbox Plus, 5 hours on Team, 24 hours on Business; support can raise it to at most 24 hours. Persistent Devboxes keep their files and can be started again. The max session shown is the Business limit.","Idle Devboxes stop automatically; the idle timeout presets are 15 minutes to 8 hours or a custom value.","Stopping keeps storage only; running processes are not described as preserved.","Egress policies can filter outbound traffic, inject credentials or proxy requests."]},"bestFor":["Coding agents that need a full dev machine with Docker, the repository checked out and a shared environment definition","Agents that must build or test on macOS (Apple Silicon with Xcode) or Windows as well as Linux","Teams already using Namespace for CI who want dev machines in the same account"],"watchOut":["Each run is capped at 3 hours on Developer and Devbox Plus (5 on Team, 24 on Business)","Low Devbox count limits: 1 per user and 3 per workspace on the pay-as-you-go plan, counting stopped ones","macOS Devboxes cost 15 times the Linux rate per size step","Team and Business plan fees do not include any Devbox usage"],"sources":[{"url":"https://namespacelabs.com/pricing","quote":"Devbox Minutes cost $0.004 each","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"Persistent storage is billed for the storage used at $0.20 per GB-month","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"50,000 Devbox Minutes","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"400GB-month storage","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"No, Devbox usage is billed separately in Devbox Minutes and follows a pay-as-you-go model by default.","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"When you sign up, you get access to the Developer plan for 30 days.","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"The limit counts both running and stopped Devboxes.","checked":"2026-11-14"},{"url":"https://namespacelabs.com/pricing","quote":"For example, a medium Linux Devbox uses 2 Devbox Minutes per minute.","checked":"2026-11-14"},{"url":"https://namespace.so/docs/devbox/lifecycle","quote":"The limit applies even while you are working in the Devbox.","checked":"2026-11-14"},{"url":"https://namespace.so/docs/devbox/lifecycle","quote":"A persistent Devbox is a machine you come back to. You create it once, work in it over days or weeks, and it stops when nobody is using it.","checked":"2026-11-14"},{"url":"https://namespace.so/docs/devbox","quote":"Linux Devboxes run from a container image you choose and include Docker.","checked":"2026-11-14"},{"url":"https://namespace.so/docs/devbox","quote":"Windows Devboxes are in early access.","checked":"2026-11-14"},{"url":"https://namespace.so/docs/security/egress-policy","quote":"Egress policies control outbound connections from your workloads.","checked":"2026-11-14"},{"url":"https://namespacelabs.com","quote":"Run untrusted or agent-generated code in isolated Mac, Windows, and Linux instances.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/namespace","score":{"dev-box":{"total":61,"rank":23,"of":38},"code-sandbox":{"total":62,"rank":35,"of":63},"ci-runner":{"total":69,"rank":6,"of":10}},"comparablePrice":{"dev-box":52.8,"code-sandbox":42.239999999999995,"ci-runner":24},"benchmarks":[{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":47445,"unit":"ms","better":"lower","date":"2026-11-13","rank":3,"of":25},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":21.79,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":2,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":11.4,"unit":"s","better":"lower","date":"2026-09-29","rank":1,"of":11}]},{"slug":"northflank","name":"Northflank","company":"Northflank","website":"https://northflank.com/product/sandboxes","docsUrl":"https://northflank.com/docs/v1/application/sandboxes/quickstart","pricingUrl":"https://northflank.com/pricing","categories":["code-sandbox","gpu"],"tagline":"MicroVM sandboxes on Northflank's container platform, billed per second for CPU and memory, on its cloud or yours.","summary":"Northflank sandboxes are Northflank services created through its API or JavaScript/Python SDKs; on Northflank Cloud, CPU sandboxes run in Cloud Hypervisor microVMs and GPU sandboxes under gVisor. Compute is billed per second against the chosen resource plan at $0.01667 per vCPU-hour and $0.00833 per GB-hour of memory, with persistent volumes and egress charged separately. Sandboxes have no time limit, can be paused to stop compute billing, and can also run on BYOC clusters in your own cloud account.","isolation":"Cloud Hypervisor microVM (CPU sandboxes); gVisor (GPU sandboxes)","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript","Python","CLI","REST API"],"regions":"Northflank Cloud regions chosen per project (GPU sandboxes need a GPU-enabled region such as asia-southeast), or BYOC on AWS, Google Cloud, Azure, Oracle, CoreWeave, Nebius and others.","features":{"pauseResume":true,"snapshots":null,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":288,"maxMemoryGib":2048},"pricing":{"rates":{"vcpuHour":0.01667,"gibHour":0.00833,"instanceHour":null,"instance":null,"storageGibMonth":0.11657,"billingIncrement":"per second","billedWhile":"Wall-clock while the sandbox has running instances, for the CPU and memory of its resource plan; paused sandboxes (scaled to zero) pay only for attached volumes.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"NVIDIA L4 24 GB","perHour":0.8},{"type":"NVIDIA A100 40 GB","perHour":1.42},{"type":"NVIDIA A100 80 GB","perHour":1.76},{"type":"NVIDIA H100 80 GB (8x only)","perHour":2.74},{"type":"NVIDIA RTX PRO 6000 96 GB","perHour":3},{"type":"NVIDIA H200 141 GB (8x only)","perHour":3.14},{"type":"NVIDIA B200 180 GB (8x only)","perHour":5.87}],"plans":[{"name":"Sandbox (free)","monthlyFee":0,"includedUsageUsd":20,"limits":"No card needed; up to $20/month of resources on Northflank Cloud across 2 services, 2 jobs and 1 database; 1 BYOC cluster up to 12 vCPU and 48 GB; limited vertical scaling, no horizontal scaling or autoscaling.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"Usage-based with no seat fees; unlimited users, deployments and workloads; GPUs, autoscaling, RBAC and the Platform SDK.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free Sandbox plan: up to $20/month of resources on Northflank Cloud (2 services, 2 jobs, 1 database), no card needed.","notes":["Rates are as published per hour: $0.01667/vCPU-hour and $0.00833/GB-hour (equal to $12/vCPU-month and $6/GB-month). Northflank labels memory in GB; we use the GB rate as the GiB rate.","Billing is per second against the configured resource plan (e.g. nf-compute-200 = 2 vCPU, 4 GB = $0.06667/hour), not measured CPU use. The pricing FAQ says billing is 'based on your configured resources and instances'; the sandbox product FAQ says 'for the CPU and memory each sandbox uses'. We follow the pricing FAQ and the plan table.","Ephemeral storage is free and included in compute. Persistent volumes cost $0.11657/GiB-month and bill whether the sandbox is running or paused; extra IOPS $7.37 per 1,000 IOPS-month and extra throughput $5.85 per 100 MiB/s-month beyond the included 3,000 IOPS and 140 MiB/s per disk.","Network egress is $0.06/GB (ingress free); an egress gateway with a dedicated IP is $15/month.","GPU prices are per GPU per hour on Northflank Cloud; H100, H200 and B200 are offered only as 8 GPUs per container. GPU plans include CPU and system RAM; the account needs GPU access.","BYOC: $0.02083 per managed node vCPU-hour (memory and storage not billed by Northflank; your cloud provider bills the infrastructure); BYOC GPUs $1.00/GB VRAM/month.","Max sandbox size is stated as 288 vCPU and '2 TB' of memory; we show 2048 GiB, although '2 TB' may be decimal.","Enterprise (custom platform fee, SLA) is not listed as a plan because no fee is published.","No sandbox-specific snapshot/checkpoint feature is described on the pages checked (volumes and S3 mounts persist state), so snapshot support is marked unknown."]},"bestFor":["Long-running or always-on agent sandboxes, since there is no time limit","Large or GPU sandboxes (up to 288 vCPU, L4 to B200 GPUs)","Teams that want the same sandbox API on Northflank Cloud or a BYOC cluster in their own cloud"],"watchOut":["Billed for the full CPU and memory of the selected plan while running, even when idle; pause the sandbox to stop compute charges","Persistent volumes bill per GiB-month even while paused; egress costs $0.06/GB","Free plan is capped at $20/month of resources, 2 services, 2 jobs and 1 database, with no horizontal scaling","A project's region cannot be changed later, and GPU sandboxes need a GPU-enabled region"],"sources":[{"url":"https://northflank.com/pricing","quote":"CPU |$0.01667 / vCPU / hour ... Memory |$0.00833 / GB / hour ... Ephemeral storage Not persisted |Free","checked":"2026-11-14"},{"url":"https://northflank.com/pricing","quote":"Persistent disk NVMe, for addon and service volumes |$0.11657 / GiB / month","checked":"2026-11-14"},{"url":"https://northflank.com/pricing","quote":"Up to $20/mo of resources on Northflank Cloud","checked":"2026-11-14"},{"url":"https://northflank.com/pricing","quote":"NVIDIA H100 80 GB VRAM","checked":"2026-11-15"},{"url":"https://northflank.com/pricing","quote":"You’ll be billed at the end of your monthly billing cycle based on your configured resources and instances, prorated to the second of usage.","checked":"2026-11-14"},{"url":"https://northflank.com/product/sandboxes","quote":"Up to 288 vCPU, 2 TB of memory, and 64 TB of volume storage, with GPUs from L4 to B200.","checked":"2026-11-14"},{"url":"https://northflank.com/product/sandboxes","quote":"No time limits Sandboxes run until you delete them, billed per second.","checked":"2026-11-14"},{"url":"https://northflank.com/docs/v1/application/sandboxes/sandboxes-on-northflank-cloud","quote":"Pausing a sandbox scales it to zero instances, which stops compute charges. Volumes attached to a paused sandbox continue to be billed, because the data is still stored.","checked":"2026-11-14"},{"url":"https://northflank.com/docs/v1/application/sandboxes/sandboxes-on-northflank-cloud","quote":"CPU sandboxes run in microVMs, and GPU sandboxes run under gVisor. ... A project's region cannot be changed after the project is created.","checked":"2026-11-14"},{"url":"https://northflank.com/pricing","quote":"$2.74 / hour","checked":"2026-11-15"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/northflank","score":{"code-sandbox":{"total":78,"rank":2,"of":63},"gpu":{"total":72,"rank":2,"of":7}},"comparablePrice":{"code-sandbox":11.73216,"gpu":2.74},"benchmarks":[]},{"slug":"novita-sandbox","name":"Novita Agent Sandbox","company":"Novita AI","website":"https://novita.ai/sandbox","docsUrl":"https://novita.ai/docs/guides/sandbox-overview","pricingUrl":"https://novita.ai/docs/guides/sandbox-pricing","categories":["code-sandbox"],"tagline":"Stateful agent sandboxes billed per second for allocated vCPU and memory, with pause, snapshots and egress rules.","summary":"Novita Agent Sandbox gives each AI agent an isolated runtime with its own filesystem, processes and network, driven from Python and JavaScript SDKs or a CLI. Compute is billed per second for allocated vCPU ($0.03528 per vCPU-hour) and memory ($0.01152 per GiB-hour) while a sandbox runs; paused sandboxes keep files and memory and pay only for storage above a 60 GB account allowance. New accounts start on a free tier capped at 5 concurrent sandboxes, and a one-time $100 promotional credit is available after a short setup survey.","isolation":"Separate runtime per sandbox with its own filesystem, process space and network stack; the underlying technology is not stated","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","CLI"],"regions":"Two US regions: us-phx-1 (default, supports secrets and snapshots) and us-virginia-1 (legacy).","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":8},"pricing":{"rates":{"vcpuHour":0.03528,"gibHour":0.01152,"instanceHour":null,"instance":null,"storageGibMonth":0.0657,"billingIncrement":"per second","billedWhile":"Allocated vCPU and memory while the sandbox is running. Paused sandboxes pay no compute, only persistent storage above the 60 GB account allowance.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free tier","monthlyFee":0,"includedUsageUsd":0,"limits":"Default for new accounts: 5 concurrent sandboxes, 1-hour maximum session, up to 2 vCPU and 4 GiB per sandbox, no priority scheduling. Usage is paid from account balance or promotional credit.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Paid tier (pay as you go)","monthlyFee":0,"includedUsageUsd":0,"limits":"Unlocks automatically once the account balance is above $0: 100 concurrent sandboxes, up to 8 vCPU and 8 GiB per sandbox, priority scheduling. Session cap is 3 hours in the docs (24 hours on the product page); long-running mode lifts it.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"One-time $100 promotional credit for new users who complete an account setup survey, no credit card required. Credits may have a validity period.","notes":["Rates converted from per-second prices: $0.0000098 per vCPU-second x 3600 = $0.03528 per vCPU-hour; $0.0000032 per GiB-second x 3600 = $0.01152 per GiB-hour.","Storage: each running sandbox has 20 GB of free ephemeral disk. Templates, snapshots and paused sandboxes use persistent storage: 60 GB free per account, then $0.00009 per GB-hour (x 730 = $0.0657 per GB-month, the storage rate we show).","Memory is set in 512 MiB steps, between 0.5 and 4 GiB per vCPU; size is fixed when the template is built.","Session length: default timeout 5 minutes. The docs cap the timeout at 1 hour (Free), 3 hours (Paid) and 4 hours (Enterprise), while the product page says 24 hours on the paid tier. Long-running mode lifts the cap, so we show no fixed maximum.","Snapshots capture filesystem and memory and are only available in the us-phx-1 region.","If the balance runs out, new sandboxes cannot start and running sandboxes are terminated with data loss.","Isolation technology (microVM, container or other) is not stated in the docs; we show it as not stated."]},"bestFor":["Agent tasks that need to pause with memory intact and resume later without paying for idle compute","Locked-down runs with outbound allowlists, private URLs and secrets injected outside the sandbox","Trying a sandbox API with a $100 credit and no card"],"watchOut":["Maximum size is 8 vCPU and 8 GiB per sandbox on the paid tier","Session limits differ between the product page (24 hours) and the docs (3 hours)","The isolation technology is not documented","Running sandboxes are terminated when the account balance runs out"],"sources":[{"url":"https://novita.ai/sandbox","quote":"$0.0000098 / vCPU-second","checked":"2026-11-13"},{"url":"https://novita.ai/sandbox","quote":"$0.0000032 / GiB-second","checked":"2026-11-13"},{"url":"https://novita.ai/sandbox","quote":"Paid tier unlocks automatically once your account balance is above $0.","checked":"2026-11-13"},{"url":"https://novita.ai/sandbox","quote":"Pay-as-you-go 100 concurrent sandboxes 24 hours max session length Up to 8 vCPU per sandbox Up to 8 GB RAM per sandbox","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-pricing","quote":"Persistent Storage for paused sandboxes | 60 GB free per account, $0.00009/GB/h for usage beyond 60 GB free storage","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-pricing","quote":"New users can complete a quick account setup survey to receive $100 in promotional credits for eligible Novita Sandbox usage. No credit card is required.","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-faq-max-runtime","quote":"Free 1 hour · Paid 3 hours · Enterprise 4 hours by default, adjustable","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-faq-cpu-memory","quote":"Up to 8 vCPU and 8 GiB of memory per sandbox on Paid (2 vCPU / 4 GiB on Free), with 20 GB of disk.","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-faq-isolation","quote":"Each sandbox has its own filesystem, process space, and network stack","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-faq-regions","quote":"Novita Sandbox currently runs in two US regions","checked":"2026-11-13"},{"url":"https://novita.ai/docs/guides/sandbox-snapshot","quote":"It captures the sandbox filesystem and memory state so you can start new sandboxes from that point later.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/novita-sandbox","score":{"code-sandbox":{"total":58,"rank":45,"of":63}},"comparablePrice":{"code-sandbox":20.52864},"benchmarks":[]},{"slug":"nvidia-openshell","name":"NVIDIA OpenShell","company":"NVIDIA","website":"https://github.com/NVIDIA/OpenShell","docsUrl":"https://docs.nvidia.com/openshell/latest/","pricingUrl":"https://github.com/NVIDIA/OpenShell","categories":["code-sandbox"],"tagline":"Open-source policy-enforced agent runtime from NVIDIA: sandboxes on Docker, Podman, Kubernetes or microVMs.","summary":"OpenShell runs autonomous agents in sandboxes where a declared policy controls file access, system calls and every outbound network connection, and credentials are injected only into requests to approved endpoints. A gateway manages sandboxes on one compute driver: Docker, Podman, Kubernetes (via the Agent Sandbox controller) or an opt-in microVM driver. Sandboxes can request GPUs, run any OCI image, and be stopped and started while keeping their workspace. It is Apache-2.0 and moving to the CNCF.","isolation":"Container per sandbox on Docker, Podman or Kubernetes with Landlock and seccomp policy enforcement; optional microVM driver (KVM or Apple Hypervisor) for a VM boundary","isolationKind":"container","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/NVIDIA/OpenShell","lastRelease":"2026-11-10","sdks":["Python","TypeScript","Go","Rust","CLI"],"regions":"Runs where you deploy the gateway: Linux, macOS on Apple Silicon, Windows with WSL 2 (experimental), or a Kubernetes cluster.","features":{"pauseResume":true,"snapshots":null,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: free software run on your own infrastructure","billedWhile":"Not applicable: you pay only for your own hardware or cluster","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"Free open-source software under the Apache-2.0 license.","notes":["Stop and start keep the sandbox's persistent workspace, policies and settings, but not process memory.","The microVM driver is opt-in and is never auto-detected; without it, sandboxes are containers.","On Kubernetes, the cluster network plugin must enforce NetworkPolicy, or sandbox pods can bypass OpenShell network policy.","The project collects anonymous telemetry by default; it can be turned off."]},"bestFor":["Running agents with per-sandbox file, syscall and network policies that need human review to widen","Keeping real API credentials out of the agent's environment","GPU agent workloads on Docker, Podman or Kubernetes"],"watchOut":["Default drivers use containers; VM isolation needs the opt-in microVM driver","Kubernetes deployments require a network plugin that enforces NetworkPolicy","Anonymous telemetry is on unless disabled"],"sources":[{"url":"https://github.com/NVIDIA/OpenShell","quote":"OpenShell is the safe, private runtime for fleets of autonomous AI agents.","checked":"2026-11-13"},{"url":"https://github.com/NVIDIA/OpenShell","quote":"Agents never see real credentials; OpenShell adds them only to requests bound for approved endpoints.","checked":"2026-11-13"},{"url":"https://github.com/NVIDIA/OpenShell","quote":"This project is licensed under the Apache License 2.0.","checked":"2026-11-13"},{"url":"https://docs.nvidia.com/openshell/latest/how-it-works/sandboxes/runtimes","quote":"When compute_driver is unset, the gateway auto-detects Kubernetes, then Podman, then Docker. The VM driver is never auto-detected.","checked":"2026-11-13"},{"url":"https://docs.nvidia.com/openshell/latest/how-it-works/sandboxes/runtimes","quote":"The MicroVM driver runs each sandbox in its own lightweight VM. It requires host virtualization: Apple Hypervisor on macOS or KVM on Linux.","checked":"2026-11-13"},{"url":"https://docs.nvidia.com/openshell/latest/how-it-works/sandboxes/overview","quote":"To request GPU resources, add --gpu","checked":"2026-11-13"},{"url":"https://docs.nvidia.com/openshell/latest/how-it-works/sandboxes/overview","quote":"Stop compute when you want to retain a sandbox and its persistent workspace without keeping its container, pod, or VM running","checked":"2026-11-13"},{"url":"https://github.com/NVIDIA/OpenShell/releases/tag/v0.1.8","quote":"v0.1.8","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/nvidia-openshell","score":{"code-sandbox":{"total":54,"rank":55,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"okteto","name":"Okteto","company":"Okteto","website":"https://www.okteto.com","docsUrl":"https://www.okteto.com/docs/","pricingUrl":"https://www.okteto.com/pricing/","categories":["dev-box"],"tagline":"Self-hosted Kubernetes platform for development, preview and agent environments; free for up to 5 users.","summary":"Okteto runs development, preview, testing and coding-agent environments as containers on a Kubernetes cluster you own, syncing local code into a remote development container. The Starter plan is free for up to 5 users with 500 environments a month; Scale and Enterprise add SSO, access control, cost controls, SLAs and BYOC, priced per user on request. The Okteto CLI is Apache-2.0, but the platform itself is commercial software installed with a license.","isolation":"Development containers in Kubernetes namespaces on your own cluster","isolationKind":"container","openSource":false,"selfHost":true,"license":null,"repoUrl":"https://github.com/okteto/okteto","lastRelease":"2026-11-05","sdks":["CLI","Okteto manifest"],"regions":"Your own Kubernetes cluster (EKS, GKE, AKS, Nutanix NKP or bare metal); BYOC on AWS or GCP on Enterprise.","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: priced per user, not per environment or minute","billedWhile":"Not applicable: you pay your own cluster costs","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Starter","monthlyFee":0,"includedUsageUsd":0,"limits":"Free for up to 5 users and 500 environments a month, one-year license, self-hosted, community forum support. No SSO, RBAC or SLA.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Starter: free self-hosted license for up to 5 users and 500 environments a month, valid for one year, no credit card needed.","notes":["Scale and Enterprise are priced per user with prices from sales.","The open-source Okteto CLI (Apache-2.0) works alone with any Kubernetes cluster but only supports the dev section of the manifest; the full platform is installed with a Helm chart and a license.","Okteto states there are no per-environment or per-minute charges."]},"bestFor":["Small teams (up to 5 users) wanting free self-hosted dev and preview environments on Kubernetes","Running coding agents in isolated environments on a cluster you already own","Developing against realistic copies of Kubernetes services"],"watchOut":["Starter is limited to 5 users and 500 environments a month and lacks SSO and RBAC","Paid tiers have no public price","Requires and runs on your own Kubernetes cluster; environments are containers"],"sources":[{"url":"https://www.okteto.com/pricing/","quote":"Okteto's free tier, for small teams: up to 5 users and 500 environments per month, with a one-year license and no credit card needed.","checked":"2026-11-15"},{"url":"https://www.okteto.com/pricing/","quote":"Okteto is priced per user account. Scale and Enterprise include unlimited environments","checked":"2026-11-15"},{"url":"https://www.okteto.com/pricing/","quote":"Okteto runs Self-Hosted on any Kubernetes cluster you own, from EKS, GKE and AKS to Nutanix NKP and bare metal.","checked":"2026-11-15"},{"url":"https://github.com/okteto/okteto","quote":"The open-source version of Okteto only supports the dev section of the Okteto manifest.","checked":"2026-11-15"},{"url":"https://github.com/okteto/okteto","quote":"This development container can use any docker image.","checked":"2026-11-15"},{"url":"https://github.com/okteto/okteto/releases/tag/3.25.0","quote":"3.25.0","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/okteto","score":{"dev-box":{"total":47,"rank":33,"of":38}},"comparablePrice":{"dev-box":null},"benchmarks":[]},{"slug":"openai-codex","name":"OpenAI Codex cloud","company":"OpenAI","website":"https://developers.openai.com/codex","docsUrl":"https://learn.chatgpt.com/docs/environments/cloud-environments","pricingUrl":"https://learn.chatgpt.com/docs/pricing","categories":["coding-agent"],"tagline":"OpenAI's coding agent running tasks in cloud VMs, included in ChatGPT Plus, Pro, Business and Enterprise plans.","summary":"Codex Cloud runs coding tasks in OpenAI-hosted VMs that keep working while your computer sleeps, started from the web, mobile, desktop app, CLI or integrations such as Slack and Linear. Each task gets its own workspace built from a reusable environment with repositories, tools, network rules and secrets. The machine has no separate price: cloud tasks draw from the same ChatGPT plan usage allowance as local Codex use, starting with the $20 Plus plan. Plus tasks get 2 vCPU / 8 GiB; Pro, Business and Enterprise get 4 vCPU / 16 GiB.","isolation":"Each cloud task runs in its own VM","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Web app","Desktop and mobile apps","CLI","Codex SDK"],"regions":"Not stated.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":4,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"No separate machine billing; cloud tasks use the plan's usage allowance","billedWhile":"Not billed by machine time. Usage depends on model, task size, context and tools, and draws from the plan's included limits and then from purchased credits.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Plus","monthlyFee":20,"includedUsageUsd":0,"limits":"Codex on the web, CLI, IDE extension and iOS, including cloud tasks. Cloud VM: 2 vCPU, 8 GiB memory, 8 GiB disk. Usage limits reset every five hours and weekly limits may apply; cloud tasks may use more of the allowance than local messages. Extra usage can be bought as credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro 100","monthlyFee":100,"includedUsageUsd":0,"limits":"Everything in Plus, with no five-hour limit at present. Cloud VM: 4 vCPU, 16 GiB memory, 32 GiB disk.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro 200","monthlyFee":200,"includedUsageUsd":0,"limits":"Everything in Plus, with no five-hour limit at present. Cloud VM: 4 vCPU, 16 GiB memory, 32 GiB disk.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro 500","monthlyFee":500,"includedUsageUsd":0,"limits":"Everything in Plus plus Ultrafast mode access, with no five-hour limit at present. Cloud VM: 4 vCPU, 16 GiB memory, 32 GiB disk.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free cloud tasks. The Free ($0) and Go ($8/month) ChatGPT plans list Codex only in the desktop app; cloud tasks start with Plus.","notes":["The cloud machine is not priced separately: there is no per-hour or per-vCPU charge. Cloud tasks count toward the same ChatGPT plan allowance as local Codex messages and ChatGPT Work, so cost depends on model, context and task size rather than machine time.","OpenAI publishes no fixed message counts for cloud tasks; its local-message estimates per five hours on Plus range from 5-45 (GPT-6 Astra) to 350-3,000 (GPT-6 Luna), and cloud tasks may use more. Pro plans currently have no five-hour limit; how much usage each Pro tier includes is not stated on the pages checked.","After the included limits, Plus and Pro users can buy credits; credits are charged per million tokens by model (for example GPT-6.1 Sol: 50 credits input, 250 credits output).","VM sizes by plan: Plus and Edu Plus 2 vCPU / 8 GiB / 8 GiB disk; Pro, Business, Enterprise, Edu and Edu Pro 4 vCPU / 16 GiB / 32 GiB disk. Larger VMs and custom specs are available on Enterprise by contract.","A task's saved VM state can be recovered for up to seven days after you last start a turn or resume it. Publishing an environment captures a prepared filesystem that new tasks start from.","Network: full public internet, a package-manager preset, or a domain allowlist per environment; network secrets are injected by a proxy only for allowed HTTPS domains. Tailscale VPN and OIDC for AWS, Google Cloud and Azure are supported.","Business, Edu and Enterprise plans are sold per seat with their own pricing; Enterprise and Edu with flexible pricing use credits instead of fixed limits. Using Codex with an API key is billed at API rates but does not include cloud tasks."]},"bestFor":["ChatGPT subscribers who want cloud coding tasks without paying for machines separately.","Teams that need per-environment network allowlists, VPN access or OIDC to cloud accounts.","Starting and following tasks from phone, web, Slack or Linear."],"watchOut":["VM size is fixed by plan: 2 vCPU / 8 GiB on Plus, 4 vCPU / 16 GiB on Pro and team plans.","Usage limits are not published as fixed numbers and are shared with all other Codex and ChatGPT Work usage.","Not available on the Free or Go plans, nor when using Codex with an API key.","Custom base images are not offered; setup is done through install scripts."],"sources":[{"url":"https://learn.chatgpt.com/docs/pricing","quote":"$20 /month","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"Each cloud task runs in a VM with these default resources:","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/pricing","quote":"Plans at $100, $200, or $500 USD per month","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/pricing","quote":"Codex on the web, in the CLI, in the IDE extension, and on iOS","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/pricing","quote":"Pro plans currently have no five-hour limit. Cloud tasks may use more of your allowance than local messages.","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/pricing","quote":"No cloud-based features (GitHub code review, Slack, etc.)","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/pricing","quote":"GPT-6 Luna at Standard speed in the desktop app, subject to rollout","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"| Plus, Edu Plus            |     2 |  8 GiB |      8 GiB |","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"| Pro, Business, Enterprise |     4 | 16 GiB |     32 GiB |","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"By default, a task's saved VM state is recoverable for up to seven days after you last start a turn or resume the task.","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"Larger VMs and custom specifications are available for Enterprise.","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"Publishing captures the prepared filesystem for new tasks.","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/pricing","quote":"Ultrafast mode access on Pro $500","checked":"2026-11-14"},{"url":"https://learn.chatgpt.com/docs/environments/cloud-environments","quote":"|Edu, Edu Pro |4 |16 GiB |32 GiB |","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/openai-codex","score":{"coding-agent":{"total":69,"rank":4,"of":11}},"comparablePrice":{"coding-agent":null},"benchmarks":[]},{"slug":"opencomputer","name":"OpenComputer","company":"OpenComputer","website":"https://opencomputer.dev","docsUrl":"https://docs.opencomputer.dev/agents/overview","pricingUrl":"https://opencomputer.dev","categories":["dev-box","coding-agent"],"tagline":"Claude Code and Codex running in cloud sandboxes, driven from Slack or an API, priced per minute of agent work.","summary":"OpenComputer runs coding agents such as Claude Code and Codex in its cloud: you mention @Computer in Slack, or deploy your own TypeScript agent and call it over HTTP. Each agent works in a sandbox with your repository, dependencies and build caches kept warm between tasks. The machine is not sold separately: plans cover a Slack workspace and include work minutes, counted only while the agent is working, and you bring your own model subscription or pay for tokens at API prices. The code is open source under Apache-2.0 and can be self-hosted for free on one VM per organization in AWS, Azure or Google Cloud.","isolation":"Not stated","isolationKind":"unknown","openSource":true,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","CLI","REST API","Slack"],"regions":"Usually the United States by default; an agent can be pinned to US, EU, Asia-Pacific and narrower regions including Canada, UK, Japan, Australia, Middle East, South America and Africa.","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Work minutes metered by the second","billedWhile":"Only while the agent is working, from picking up a message to its reply; idle threads and time waiting on people are free.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pro","monthlyFee":20,"includedUsageUsd":0,"limits":"2,000 work minutes a month for the whole Slack workspace. Extra minutes $0.01 each from prepaid credits. Model use on your own subscription or key, or on OpenComputer's tokens at API prices.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":200,"includedUsageUsd":0,"limits":"25,000 work minutes a month for the whole Slack workspace. Extra minutes $0.01 each from prepaid credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"3-day free trial with 600 work minutes and $10 of model tokens, no card; it ends at 600 minutes or day 3, whichever comes first.","notes":["The underlying machine has no separate hourly price. Work minutes beyond a plan cost $0.01 each, which is $0.60 per hour of agent work; two threads working at once count twice.","Serverless agents you deploy share the plan's work minutes with @Computer in Slack.","Credit packs are $10, $50 or $200 with optional auto top-up; nothing is billed after the fact.","Enterprise (over 60,000 work minutes a month) has volume pricing by contact.","Self-hosting is free under the Apache-2.0 license: it runs on one VM per organization in your AWS, Azure or Google Cloud account, and you pay your cloud and model provider directly.","Sandbox size, isolation technology and persistence details are not published on the pages checked."]},"bestFor":["Teams that want to hand small code changes to an agent from a Slack thread","Using an existing Claude or ChatGPT subscription for cloud agent work","Deploying custom agents as HTTP endpoints, on schedules or webhooks","Teams that want to run the same Slack agent free on their own cloud account"],"watchOut":["No general-purpose machine or SSH dev box is sold; access is through the agent","Sandbox size, isolation and limits are not published","Plans are per Slack workspace, and parallel threads use minutes in parallel"],"sources":[{"url":"https://opencomputer.dev","quote":"Free, on your own cloud. Apache-2.0. You pay your cloud and your model provider, not us.","checked":"2026-11-14"},{"url":"https://docs.opencomputer.dev/self-hosting/overview","quote":"You create one VM per organization in your cloud account","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"Pro, 2,000 minutes included","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"25,000 work minutes a month","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"$200 / month","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"$0 for the model, on your Claude or ChatGPT plan","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"Past your plan's minutes, each extra minute is $0.01, paid from prepaid credits","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"No card. Includes $10 of our model tokens. Ends at 600 minutes or day 3, whichever comes first.","checked":"2026-11-14"},{"url":"https://opencomputer.dev","quote":"From the moment @ Computer picks up your message to its reply: reading code, running commands, builds and tests. Metered by the second.","checked":"2026-11-14"},{"url":"https://docs.opencomputer.dev/agents/regions.md","quote":"By default, an agent's sandbox runs wherever capacity is available, usually in the United States.","checked":"2026-11-14"},{"url":"https://docs.opencomputer.dev/agents/egress.md","quote":"By default, an agent's sandbox can reach any address on the internet.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/opencomputer","score":{"dev-box":{"total":41,"rank":37,"of":38},"coding-agent":{"total":35,"rank":11,"of":11}},"comparablePrice":{"dev-box":null,"coding-agent":null},"benchmarks":[]},{"slug":"orgo","name":"Orgo","company":"Orgo","website":"https://www.orgo.ai","docsUrl":"https://docs.orgo.ai/","pricingUrl":"https://docs.orgo.ai/guides/pricing","categories":["dev-box","browser"],"tagline":"Persistent Linux and Windows desktop VMs for computer-use agents, by the hour or as monthly plan capacity.","summary":"Orgo provides full desktop computers for AI agents: each Linux or Windows computer is its own virtual machine with a disk, desktop, browser and terminal, controlled through screenshots, clicks, typing and bash over an API, SDKs or MCP. Computers run until you stop them, with no time limit, and are backed up daily. Linux computers can run by the hour from credit, from $0.0548 an hour for a Medium (2 vCPU, 8 GB), or on a monthly plan that includes shared vCPU and RAM. Windows runs only on a plan; Mac and iPhone hardware is separate.","isolation":"Isolated virtual machine per computer","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI","MCP server","REST API"],"regions":"Not stated","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":6,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.0547945205479452,"instance":"Medium Linux: 2 vCPU, 8 GB RAM, 40 GB disk","storageGibMonth":0.05,"billingIncrement":"Per second, no hourly minimum","billedWhile":"By the hour: the running price covers vCPU, RAM and disk while running; a stopped computer pays only for its disk. Plan capacity is a flat monthly charge.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":0.023116438356164382},"gpu":[],"plans":[{"name":"Starter","monthlyFee":29,"includedUsageUsd":5,"limits":"2 vCPU and 8 GB of shared capacity for computers on the plan, $5 of monthly AI & Platform credit, 1 GB residential IP data, 5 people per workspace, up to 50 computers.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Growth","monthlyFee":99,"includedUsageUsd":10,"limits":"8 vCPU and 32 GB of shared capacity, $10 of monthly AI & Platform credit, 5 GB residential IP data, 5 people per workspace, up to 50 computers.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":399,"includedUsageUsd":50,"limits":"36 vCPU and 128 GB of shared capacity, $50 of monthly AI & Platform credit, 20 GB residential IP data, 25 people per workspace, Windows included, up to 50 computers.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Free account: $5 starting credit (or $2 for an agent signup) that runs one Medium Linux computer by the hour.","notes":["Hourly price = the size's monthly plan price plus 25%, divided by 730 hours. Medium: $32 x 1.25 / 730 = $0.0547945 an hour (shown by Orgo as $0.055). Other sizes: Small 0.5 vCPU / 4 GB $0.023, Large 4 / 16 GB $0.11, XL 6 / 32 GB $0.202. Prices are not linear in one unit.","The browser-hour price we show is the Small size ($13.50 x 1.25 / 730 = $0.0231164 an hour), the cheapest Linux desktop; Small is API-only and on a free account only Medium runs by the hour.","Plan capacity added beyond a plan costs $2.50 per GB of RAM, $5 per vCPU and $1 per 20 GB of disk a month; whole sizes are Small $13.50, Medium $32, Large $64, XL $118 a month. Plan allowances are reserved capacity, not hours.","Disk is $1 per 20 GB block a month ($0.05 per GB-month); a stopped Medium pays $0.0027 an hour.","XL (6 vCPU / 32 GB) is the largest self-serve size; 2XL (8 vCPU / 64 GB) is by request. Windows computers cost $15 a month extra below Scale and are not offered by the hour.","Residential IP data beyond a plan's allowance is $3 per GB. Orgo states memory in GB without saying GB or GiB."]},"bestFor":["Computer-use agents that need a full Linux or Windows desktop rather than a browser tab","Long-running agent desktops with no session time limit","Watching or taking over an agent's screen live"],"watchOut":["Hourly use is 25% dearer than plan capacity for always-on computers","Windows needs a plan and costs extra below Scale","Free accounts run one Medium computer at a time and other sizes need a plan","Regions and network controls are not documented on the pages checked"],"sources":[{"url":"https://docs.orgo.ai/guides/pricing","quote":"A Medium is $32 × 1.25 / 730 = $0.055 an hour.","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/pricing","quote":"| Small | 0.5 | 4 GB | 20 GB | $0.023/hr | $0.0014/hr |","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/pricing","quote":"Time is billed by the second, with no hourly minimum.","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/pricing","quote":"| Disk | $1 per 20 GB | $9.60 per 20 GB |","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/billing","quote":"| Starter | $29 | $276 | 2 vCPU, 8 GB | 1 GB | 5 |","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/billing","quote":"Each plan includes monthly AI & Platform credits: $5 on Starter, $10 on Growth, $50 on Scale.","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/billing","quote":"Starting credit ($5, or $2 for an agent signup) runs one hourly computer","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/guides/instance-types","quote":"On your own, one computer goes up to XL: 6 vCPU and 32 GB RAM.","checked":"2026-11-15"},{"url":"https://docs.orgo.ai/llms.txt","quote":"Every Linux and Windows computer is its own isolated virtual machine with its own disk, a full desktop, a browser, and a terminal","checked":"2026-11-15"},{"url":"https://www.orgo.ai/pricing","quote":"Runs until you stop it No time limit.","checked":"2026-11-15"},{"url":"https://www.orgo.ai/pricing","quote":"Linux and Windows computers are backed up every day.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-13","page":"https://aisandboxhunt.com/orgo","score":{"dev-box":{"total":60,"rank":26,"of":38},"browser":{"total":78,"rank":2,"of":12}},"comparablePrice":{"dev-box":36.054794520547944,"browser":0.023116438356164382},"benchmarks":[]},{"slug":"podflare","name":"Podflare","company":"Podflare","website":"https://podflare.ai","docsUrl":"https://docs.podflare.ai/","pricingUrl":"https://podflare.ai/pricing","categories":["code-sandbox"],"tagline":"KVM microVM sandboxes with a persistent Python REPL and fast fork, in private alpha with a $200 starting credit.","summary":"Podflare runs each sandbox as a KVM-backed microVM with its own kernel, root access, outbound internet and a Python REPL whose state carries across calls. Sandboxes can be forked into copy-on-write children and frozen to disk as Persistent Spaces. Compute is metered per second at $0.05 per vCPU-hour and $0.016 per GiB-hour plus $0.09 per GB of egress, and monthly tiers (Free, Pro $50, Scale $200) set sandbox size, lifetime and concurrency. The service is labelled private alpha, and new accounts get $200 of credit.","isolation":"KVM-backed microVM per sandbox with a dedicated guest kernel","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","MCP server","REST API"],"regions":"The pricing page lists EU and US-West; the docs list five regions (us-west, us-central, us-east, eu, sg), routed to the nearest by default with optional region pinning.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":null,"networkControls":true,"maxSessionHours":24,"maxVcpu":8,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":0.05,"gibHour":0.016,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per second, billed monthly","billedWhile":"Pay-as-you-go compute for usage beyond the tier's base capacity; the pricing page does not say whether idle sandboxes inside the warm window are billed.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1 GB RAM, 2 vCPU, 4 GB disk per sandbox; 5-minute idle timeout; 30-minute maximum lifetime; 10 concurrent sandboxes.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":50,"includedUsageUsd":0,"limits":"4 GB RAM, 4 vCPU, 16 GB disk per sandbox; 30-minute idle timeout; 8-hour maximum lifetime; 50 concurrent sandboxes; 24-hour support response. Includes base capacity of unstated size.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":200,"includedUsageUsd":0,"limits":"16 GB RAM, 8 vCPU, 64 GB disk per sandbox; 2-hour idle timeout; 24-hour maximum lifetime; 500 concurrent sandboxes; 4-hour support response. Includes base capacity of unstated size.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"New accounts get $200 of credit, which applies to both subscription and usage fees; no card hold during the private alpha.","notes":["Pay-as-you-go rates: $0.05 per vCPU-hour, $0.016 per GiB RAM-hour and $0.09 per GB of egress, metered by the second and billed monthly.","Pro and Scale 'include your base capacity' and roll overage to pay-as-you-go; the amount of included usage is not stated, so we show none.","Storage pricing for Persistent Spaces is not stated.","Sizes are clamped to the tier: the product page lists 2-16 vCPU and 1-16 GB RAM overall, but the largest self-serve tier (Scale) allows 8 vCPU / 16 GB. Enterprise offers custom sizes and unlimited session length.","Paid upgrades are activated by email during the private alpha. SOC 2 Type II is in progress, not certified.","Outbound internet is on by default; it can be turned off per sandbox (egress=False). Finer allowlists are not described.","Region list differs between the pricing page (EU, US-West) and the docs (five regions)."]},"bestFor":["Tree-search or multi-branch agents that fork a running sandbox into many copies","Code-interpreter tools that need Python state kept across calls","Prototyping on a $200 starting credit"],"watchOut":["Private alpha: APIs and tier policy may still change, and paid upgrades are activated by email","Free tier sandboxes stop after 30 minutes; Scale tops out at 24 hours","Egress is billed at $0.09 per GB","SOC 2 is not yet certified"],"sources":[{"url":"https://podflare.ai/pricing","quote":"$0.05 / vCPU-hour · $0.016 / GiB RAM-hour · $0.09 / GB egress. Metered by the second; billed monthly.","checked":"2026-11-13"},{"url":"https://podflare.ai/pricing","quote":"New accounts get $200 in credit — enough to prototype a full agent integration. Credit applies to both subscription and usage fees.","checked":"2026-11-13"},{"url":"https://podflare.ai/pricing","quote":"Pro and Scale include your base capacity. Overage beyond tier limits rolls to pay-as-you-go compute at published rates.","checked":"2026-11-13"},{"url":"https://podflare.ai/pricing","quote":"16 GB RAM • 8 vCPU • 64 GB disk per sandbox","checked":"2026-11-13"},{"url":"https://podflare.ai/pricing","quote":"Private alpha paid upgrades activated by email within 15–30 min — no card hold","checked":"2026-11-13"},{"url":"https://podflare.ai","quote":"KVM-backed virtual machine with its own kernel and page tables.","checked":"2026-11-13"},{"url":"https://docs.podflare.ai/","quote":"Podflare is in private alpha.","checked":"2026-11-13"},{"url":"https://docs.podflare.ai/","quote":"5 regions (us-west, us-central, us-east, eu, sg)","checked":"2026-11-13"},{"url":"https://docs.podflare.ai/","quote":"Opt out per-sandbox with `egress=False` for untrusted code.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/podflare","score":{"code-sandbox":{"total":71,"rank":15,"of":63}},"comparablePrice":{"code-sandbox":28.864},"benchmarks":[]},{"slug":"promigence","name":"Promigence","company":"Promigence","website":"https://www.promigence.ai","docsUrl":"https://www.promigence.ai/docs","pricingUrl":"https://www.promigence.ai/pricing","categories":["code-sandbox"],"tagline":"Sandboxes for agents, evals and CI from verified snapshots, billed per second with a quote and hard spend cap per run.","summary":"Promigence starts sandboxes from snapshots that must pass their own verify command at build time, and forks many identical copies at once for agent runs, evaluations, RL rollouts and CI jobs. Usage is metered per second at $0.0504 per vCPU-hour and $0.0162 per GiB-hour, paused time is not billed, and snapshot storage and egress are included. Every run is priced before it starts and held to a hard spend cap, and runs report environment failures separately from task failures. New accounts get $50 of free credit; paid plans are arranged by email.","isolation":"Isolated sandbox per run; the isolation technology is not stated","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","Python","CLI","MCP server","REST API"],"regions":"Not stated on the pages checked; the API reports a location code such as us-east.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":0,"billingIncrement":"per second (1-second minimum)","billedWhile":"From hand-over of a ready sandbox until it ends. Paused time is not billed, and nothing accrues before the sandbox is handed over.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Pay as you go at list rates after the $50 starting credit. 20 sandboxes at once, sessions up to 1 hour, sizes up to Large (8 vCPU), 20 GB of data out a month.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":50,"includedUsageUsd":0,"limits":"Set up by email. 'More compute each month than the plan costs' (amount not stated), up to 100 sandboxes at once, verified snapshot builds, harness adapters, email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":250,"includedUsageUsd":0,"limits":"Set up by email. 5x the usage of Pro (or 20x for $500), up to 500 sandboxes at once (1,000 at $500), capacity held warm, shared Slack channel; usage past the allowance below list by commitment.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"$50 of free credit to start; the owner claims it once in the browser.","notes":["Sizes are fixed tiers at the same per-resource rate: Small 2 vCPU / 4 GiB ($0.1656/hour), Medium 4 vCPU / 16 GiB ($0.4608/hour), Large 8 vCPU / 32 GiB ($0.9216/hour). The list prices are rounded to $0.17, $0.46 and $0.92.","Snapshot storage and egress are included (we show storage as $0); the trial caps data out at 20 GB a month.","Verified snapshot builds bill at their size's rate for the time they run, at most an hour; builds that fail on Promigence's side are not billed.","Paused sandboxes are kept 30 days and do not count against concurrency.","Plans are set up by email; the dollar value of usage included with Pro and Max is not stated, so we show none.","Free accounts can run sessions up to an hour (15 minutes by default); the limit for paid plans is not stated.","The isolation technology (microVM, container or other) is not described; the security page only says no sandbox can reach another.","Outbound network can be limited with an allowlist or turned off. GPUs are not offered."]},"bestFor":["Evaluation and RL runs that fork many identical sandboxes and need environment failures kept apart from agent failures","Teams that want a price quote and a hard spend cap before each batch starts","Running GitHub Actions or GitLab CI jobs in sandboxes from the same account"],"watchOut":["Free accounts are limited to 1-hour sessions and 20 sandboxes at once","Paid plans are arranged by email and their included usage is not stated in dollars","The isolation technology is not documented","Some accounts must answer setup questions before starting work"],"sources":[{"url":"https://www.promigence.ai/pricing","quote":"Metered per second at $0.0504 per vCPU-hour plus $0.0162 per GiB-hour","checked":"2026-11-13"},{"url":"https://www.promigence.ai/pricing","quote":"$0.000046 per second ($0.1656 per hour)","checked":"2026-11-13"},{"url":"https://www.promigence.ai/pricing","quote":"Plans are set up with you by email.","checked":"2026-11-13"},{"url":"https://www.promigence.ai/pricing","quote":"20 sandboxes at once, sessions up to an hour","checked":"2026-11-13"},{"url":"https://www.promigence.ai/pricing","quote":"Snapshot storage: Included. Included in every plan.","checked":"2026-11-13"},{"url":"https://www.promigence.ai/pricing","quote":"A paused environment accrues nothing, and nothing accrues before one is handed to you.","checked":"2026-11-13"},{"url":"https://www.promigence.ai/pricing","quote":"5× the usage of Pro, or 20× for $500","checked":"2026-11-13"},{"url":"https://www.promigence.ai/docs/limits","quote":"A new account runs 20 sandboxes at once, each for up to an hour (15 minutes unless `--timeout` asks for more), on sizes up to large (8 vCPU), with 20 GB of data out a month.","checked":"2026-11-13"},{"url":"https://www.promigence.ai/docs/limits","quote":"It is kept for 30 days, and resuming it starts the 30 days again.","checked":"2026-11-13"},{"url":"https://www.promigence.ai/security","quote":"No sandbox can see, reach or address another, yours or anyone else's.","checked":"2026-11-13"},{"url":"https://www.promigence.ai","quote":"A new account's owner claims its free credit once, in the browser","checked":"2026-11-13"},{"url":"https://www.promigence.ai/docs/llms-full.txt","quote":"There are no GPUs or Windows images.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/promigence","score":{"code-sandbox":{"total":55,"rank":52,"of":63}},"comparablePrice":{"code-sandbox":29.1456},"benchmarks":[]},{"slug":"railway-sandboxes","name":"Railway Sandboxes","company":"Railway","website":"https://railway.com/sandboxes","docsUrl":"https://docs.railway.com/sandboxes","pricingUrl":"https://railway.com/pricing","categories":["code-sandbox"],"tagline":"Isolated Linux VMs inside a Railway project, billed per second for CPU used and memory in use.","summary":"Railway Sandboxes are ephemeral Linux VMs scoped to a Railway environment, created from the TypeScript SDK, CLI, dashboard or GraphQL API. They boot Ubuntu with git, Node and common coding agents preinstalled, and can join the project's private network to reach its databases and services. CPU is billed for seconds actually used and memory for what is in use, so an idle sandbox still pays for memory. Checkpoints save the disk to boot or fork later.","isolation":"Isolated Linux VM per sandbox","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","CLI","GraphQL API","Dashboard"],"regions":"US West (default), US East, EU West and Southeast Asia.","features":{"pauseResume":null,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":24,"maxMemoryGib":24},"pricing":{"rates":{"vcpuHour":0.069444,"gibHour":0.069444,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per second","billedWhile":"While the sandbox runs: CPU for active use only, memory in use (including OS and file cache) even while idle, and outbound traffic. Destroying the sandbox stops compute charges.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":1,"limits":"$1 of usage credit per month. 10 running sandboxes per environment, max 2 vCPU / 2 GB per sandbox, idle timeout up to 5 minutes.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Hobby","monthlyFee":5,"includedUsageUsd":5,"limits":"Includes $5 of usage per month. 50 running sandboxes per environment, max 8 vCPU / 8 GB, idle timeout up to 2 hours or disabled.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":20,"includedUsageUsd":20,"limits":"Includes $20 of usage per month. 100 running sandboxes per environment, max 24 vCPU / 24 GB (the pricing page says 32 vCPU / 32 GB), idle timeout up to 2 hours or disabled.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan with $1 of usage credit each month, after a 30-day trial with a one-time $5 credit. Sandboxes are available on Free and the trial, capped at 2 vCPU / 2 GB.","notes":["Rates are $0.00001929 per vCPU-second (active use) and $0.00001929 per GB-second (memory in use), multiplied by 3,600; Railway quotes about $50 per vCPU-month and $50 per GB-month. Railway writes GB, which we treat as GiB.","Sandbox rates are higher than Railway's regular service rates ($0.00000772 per vCPU-second and $0.00000386 per GB-second).","Egress is $0.05 per GB. A price for checkpoint storage is not stated.","Plan fees count toward usage: Hobby and Pro include usage equal to the fee, and only usage above it is billed.","Maximum size conflict: the sandboxes page lists 24 vCPU / 24 GB on Pro, the pricing page 32 vCPU / 32 GB. We show 24.","Checkpoints and forks copy the disk only; each new sandbox boots fresh, without processes or memory.","Networking is ISOLATED (outbound internet only, default) or PRIVATE (joins the project's private network); public domains need PRIVATE."]},"bestFor":["Agents that need to reach a project's existing databases and services over a private network","Short tasks with low average CPU use, since only busy CPU seconds are billed","Trying several agent attempts from one prepared checkpoint"],"watchOut":["Memory in use keeps billing while a sandbox sits idle","Sandbox CPU and memory rates are well above Railway's normal service rates","Free and trial sandboxes are capped at 2 vCPU / 2 GB with a 5-minute idle timeout","No pause: checkpoints restore files into a new sandbox, not running processes"],"sources":[{"url":"https://railway.com/sandboxes","quote":"| Memory | $0.00001929 per GB/s (in use) | $50 per GB |","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"| CPU | $0.00001929 per vCPU/s (active use) | $50 per vCPU |","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"Idle sandboxes remain billable.","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"| Pro | 100 | 8 vCPU / 8 GB | 24 vCPU / 24 GB | 30 min | Up to 120 min, or never |","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"Regions: US West (default), US East, EU West, Southeast Asia.","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"Each new sandbox boots fresh; processes and memory aren't copied.","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"Networking: `ISOLATED` (default, outbound internet only) or `PRIVATE` (joins the environment's private network).","checked":"2026-11-13"},{"url":"https://railway.com/sandboxes","quote":"A Railway sandbox is an isolated, ephemeral Linux VM scoped to a Railway environment.","checked":"2026-11-13"},{"url":"https://railway.com/pricing","quote":"Start with a 30-day free trial with $5 credits, then $1 per month","checked":"2026-11-13"},{"url":"https://railway.com/pricing","quote":"32 vCPU / 32 GB","checked":"2026-11-13"},{"url":"https://railway.com/pricing","quote":"$0.05 per GB","checked":"2026-11-13"},{"url":"https://docs.railway.com/reference/pricing/plans","quote":"For running small apps with $1 of free credit per month","checked":"2026-11-13"},{"url":"https://docs.railway.com/reference/pricing/plans","quote":"The Hobby plan includes $5 of resource usage per month.","checked":"2026-11-13"},{"url":"https://docs.railway.com/reference/pricing/plans","quote":"the Pro plan includes $20 of resource usage per month","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/railway-sandboxes","score":{"code-sandbox":{"total":59,"rank":42,"of":63}},"comparablePrice":{"code-sandbox":56.2218624},"benchmarks":[]},{"slug":"render","name":"Render Sandboxes","company":"Render","website":"https://render.com/sandboxes","docsUrl":"https://render.com/docs/sandboxes","pricingUrl":"https://render.com/pricing","categories":["code-sandbox","dev-box"],"tagline":"Firecracker microVMs on Render that suspend when idle, billed for active CPU time plus provisioned memory.","summary":"Render Sandboxes run each sandbox in its own Firecracker microVM, created and driven from the Render CLI, the TypeScript or Python SDK, or the API. CPU is billed at $0.12 per active CPU hour and memory at a fixed hourly rate for the compute plan's RAM while the sandbox runs; a suspended sandbox has no compute charge. By default a sandbox suspends after 15 minutes without commands or file transfers and resumes on the next one, and both the idle and runtime timeouts can be turned off so a sandbox runs indefinitely. The product is in public beta and runs in one region (Oregon).","isolation":"Firecracker microVM per sandbox","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","CLI","REST API"],"regions":"1 region during the beta: Oregon.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":0.12,"gibHour":0.019,"instanceHour":null,"instance":null,"storageGibMonth":0.08,"billingIncrement":"prorated to the second (Render's general compute billing rule; no separate increment stated for sandboxes)","billedWhile":"CPU: only active CPU time while the sandbox runs. Memory: the compute plan's full provisioned RAM while the sandbox runs. Nothing for compute while suspended; snapshot storage billed separately.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Hobby","monthlyFee":0,"includedUsageUsd":0,"limits":"No workspace fee; sandbox compute billed at usage rates. 100 concurrent sandboxes by default, 1 seat, no maximum session length.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":25,"includedUsageUsd":0,"limits":"$25/month workspace fee plus compute. 1,000 concurrent sandboxes by default (up to 5,000 on request), unlimited seats, no maximum session length.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":499,"includedUsageUsd":0,"limits":"$499/month workspace fee plus compute. 1,000 concurrent sandboxes by default (up to 10,000 on request), multiple workspaces, SAML SSO, no maximum session length.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"none","freeTier":"No free sandbox usage is stated. The Hobby workspace has no monthly fee, but sandbox compute is billed at the usage rates; Render's free compute plans cover web services, Key Value and Postgres only.","notes":["Memory is priced per compute plan: $0.038/hour for 2 GB, $0.076 for 4 GB, $0.152 for 8 GB and $0.304 for 16 GB, which is $0.019 per GB-hour on every plan. Render writes GB without saying GB or GiB; we use the GB-hour rate as the GiB-hour rate.","Compute plans are fixed shapes: 1 CPU / 2 GB (default), 2 CPU / 4 GB, 4 CPU / 8 GB, 8 CPU / 16 GB. CPU is billed at $0.12 per active CPU hour on every plan.","Snapshot storage is $0.08 per GB per month. Each sandbox includes 40 GB of writable disk; no separate disk charge is listed.","Timeouts: the runtime timeout defaults to 24 hours (range 1 minute to 30 days, or 0 to disable) and restarts on each resume; the idle timeout defaults to 15 minutes (5 minutes to 24 hours, or 0 to disable). A suspended sandbox can be resumed for 30 days. Background processes do not keep a sandbox active after the command that started them ends.","Termination permanently deletes the sandbox filesystem; capture a snapshot first to keep state.","Wording differs between pages: the docs introduction calls sandboxes 'managed, ephemeral environments', while the product page says a suspended sandbox resumes 'with its files, memory, and running processes intact'.","Concurrency figures differ by page: the pricing table gives 100 concurrent sandboxes on Hobby and 1,000 on Pro and Scale, while the limits page gives 1,000 concurrent sandboxes per sandbox group and up to 10 groups per workspace.","Public beta: Render says APIs, defaults and limits might change. Beta limitations include no private networking to other Render services and no public URLs for apps inside a sandbox; credential brokering and scoped API keys are early access.","Workspace plan fees cover the whole Render workspace, not only sandboxes. Outbound bandwidth is included up to the plan allowance, then $0.15 per GB."]},"bestFor":["Agent work that waits on model or API calls, since idle CPU is not billed and idle sandboxes suspend.","Teams already hosting on Render that want sandboxes in the same workspace and billing.","Long-lived agent environments: timeouts can be turned off and suspended sandboxes resume with their state."],"watchOut":["Public beta, with one region (Oregon) and limits that may change.","Memory is billed for the plan's full RAM while the sandbox runs, even when idle.","Sizes are fixed plans up to 8 CPU / 16 GB; no GPU is described.","No public URLs for apps running inside a sandbox during the beta."],"sources":[{"url":"https://render.com/pricing","quote":"Billing for active CPU time + provisioned memory","checked":"2026-11-14"},{"url":"https://render.com/pricing","quote":"$0.12 / active CPU hour | 2 GB RAM | $0.038/hour for provisioned memory | 1c-2g","checked":"2026-11-14"},{"url":"https://render.com/pricing","quote":"$0.12 / active CPU hour | 16 GB RAM | $0.304/hour for provisioned memory | 8c-16g","checked":"2026-11-14"},{"url":"https://render.com/pricing","quote":"Snapshot storage for $0.08/GB per month","checked":"2026-11-14"},{"url":"https://render.com/pricing","quote":"Pro | $25/month + compute","checked":"2026-11-14"},{"url":"https://render.com/pricing","quote":"Default limit of 1,000 concurrent sandboxes on Pro, Scale workspaces","checked":"2026-11-14"},{"url":"https://render.com/pricing","quote":"billing is prorated to the second","checked":"2026-11-14"},{"url":"https://render.com/sandboxes","quote":"Each Render sandbox runs in its own Firecracker microVM","checked":"2026-11-14"},{"url":"https://render.com/sandboxes","quote":"While a sandbox runs, you pay for the CPU it actively uses and the memory in its compute plan. A suspended sandbox has no compute charges.","checked":"2026-11-14"},{"url":"https://render.com/docs/sandboxes","quote":"Render Sandboxes is in public beta.","checked":"2026-11-14"},{"url":"https://render.com/docs/sandboxes-limits","quote":"1 (Oregon)","checked":"2026-11-14"},{"url":"https://render.com/docs/sandboxes-limits","quote":"40 GB","checked":"2026-11-14"},{"url":"https://render.com/docs/sandboxes-creating","quote":"For a new sandbox, defaults to `900` (15 minutes)","checked":"2026-11-14"},{"url":"https://render.com/docs/sandboxes-creating","quote":"To run a sandbox indefinitely, set both of these timeouts to `0`.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/render","score":{"code-sandbox":{"total":63,"rank":32,"of":63},"dev-box":{"total":63,"rank":20,"of":38}},"comparablePrice":{"code-sandbox":26.048,"dev-box":27.279999999999998},"benchmarks":[]},{"slug":"replit","name":"Replit","company":"Replit","website":"https://replit.com","docsUrl":"https://docs.replit.com/","pricingUrl":"https://replit.com/pricing","categories":["dev-box","coding-agent"],"tagline":"Browser-based workspace where Replit Agent builds, runs and publishes apps; the machine is included in a plan.","summary":"Replit is a browser-based development platform built around Replit Agent, which builds apps from prompts in a hosted cloud workspace and publishes them with databases and storage. The development machine is not priced on its own: plans include agent usage and monthly credits, and publishing, databases and storage are billed from those credits. Core costs $20 a month with $20 of credit for more powerful models; Pro starts at $100 a month with $100 of credits and up to 15 collaborators. The free Starter plan gives daily agent credits and one published app.","isolation":"Not stated","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Web IDE","Mobile app"],"regions":"Not stated","features":{"pauseResume":null,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not stated","billedWhile":"The development workspace is included in the plan; publishing, databases, storage and paid agent modes draw on credits.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Starter","monthlyFee":0,"includedUsageUsd":0,"limits":"Daily agent credits up to a monthly cap, monthly credits for databases, storage and publishing, Lite build only, 1 published app that goes down after 30 days.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Core","monthlyFee":20,"includedUsageUsd":20,"limits":"$20 a month ($18 billed yearly). Up to 30 hours of chat and 60 projects on Free Mode, $20 toward more powerful models, unlimited published apps, up to 5 collaborators, 1 active background task. Pay-as-you-go beyond credits.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":100,"includedUsageUsd":100,"limits":"$100 a month ($90 billed yearly) with $100 of credits; higher credit tiers up to $2,000 a month. Up to 15 collaborators, 10 parallel tasks, credit rollover for two months, 28-day database rollback, premium support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Starter plan: free daily agent credits up to a monthly cap and monthly credits for cloud usage, with one published app.","notes":["Replit does not sell its development machine by the hour; there is no compute rate to compare. Plans bundle agent use, credits and the workspace.","Pro yearly tiers listed: $90, $215, $425, $825 and $2,000 a month.","Workspace CPU, memory, isolation and regions are not stated on the pricing and plan pages checked.","Enterprise adds SSO/SCIM, single-tenant and private networking options at custom pricing."]},"bestFor":["Building and publishing apps from prompts without setting up a machine","Non-developers and small teams prototyping in the browser","Projects that want hosting, databases and the editor in one place"],"watchOut":["No raw machine to rent; workspace resources are not published","Spending is driven by agent credits, which can run out quickly on heavy use","Not an API sandbox for running your own agents","Free published apps go down after 30 days"],"sources":[{"url":"https://replit.com/pricing","quote":"$20 towards more powerful models","checked":"2026-11-14"},{"url":"https://replit.com/pricing","quote":"Up to 30 hours of chat on Free Mode","checked":"2026-11-14"},{"url":"https://replit.com/pricing","quote":"$100 towards most powerful models","checked":"2026-11-14"},{"url":"https://replit.com/pricing","quote":"$90 / month $215 / month $425 / month $825 / month $2,000 / month","checked":"2026-11-14"},{"url":"https://docs.replit.com/billing/plans/replit-core","quote":"| Active background tasks | 0 | 1 | 10 |","checked":"2026-11-14"},{"url":"https://docs.replit.com/billing/plans/replit-core","quote":"| Collaboration seats | 0 | 5 | 15 |","checked":"2026-11-14"},{"url":"https://docs.replit.com/billing/plans/starter-plan","quote":"Starter plan users receive daily credits for Agent usage, up to a monthly cap.","checked":"2026-11-14"},{"url":"https://docs.replit.com/billing/plans/starter-plan","quote":"This published link will automatically go down after 30 days.","checked":"2026-11-14"},{"url":"https://docs.replit.com/billing/plans/replit-pro","quote":"Unused credits roll over for two months","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/replit","score":{"dev-box":{"total":38,"rank":38,"of":38},"coding-agent":{"total":36,"rank":10,"of":11}},"comparablePrice":{"dev-box":null,"coding-agent":null},"benchmarks":[]},{"slug":"riza","name":"Riza Code Interpreter API","company":"Riza","website":"https://riza.io","docsUrl":"https://docs.riza.io/","pricingUrl":"https://riza.io/pricing","categories":["code-sandbox"],"tagline":"API that runs untrusted Python, JavaScript, TypeScript, Ruby or PHP in a WebAssembly sandbox, priced per request.","summary":"Riza runs short scripts or functions sent over an API in a fresh WebAssembly sandbox per execution, with no filesystem or network access unless you grant it. It supports Python, JavaScript, TypeScript, Ruby and PHP, with third-party packages for Python and JavaScript through custom runtimes. Pricing is by request count, not compute time: Hobby is free for up to 100,000 requests a month and Pro is $250 a month for 1,000,000. Enterprise adds self-hosting.","isolation":"WebAssembly sandbox per execution; modules are not shared across executions","isolationKind":"isolate","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","Node.js","Go","REST API","MCP server"],"regions":"Not stated","features":{"pauseResume":false,"snapshots":false,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":0.25,"maxVcpu":null,"maxMemoryGib":0.125},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per request (each code execution)","billedWhile":"Each execution request counts against the plan's monthly request quota; execution time and resources are not metered separately.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Hobby","monthlyFee":0,"includedUsageUsd":0,"limits":"Up to 100,000 requests and 100 GB data transfer per month; 30-second execution limit per request; 3 custom runtimes.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Pro","monthlyFee":250,"includedUsageUsd":0,"limits":"1,000,000 requests and 500 GB data transfer per month; up to 15-minute execution limit per request; 50 custom runtimes. Higher limits use custom usage-based pricing.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Hobby plan is free with up to 100,000 requests per month, 30-second executions and 3 custom runtimes.","notes":["There is no per-hour compute price, so hourly rates are left empty. Pro works out to $0.00025 per request at its full 1,000,000-request quota.","Overage pricing above the Pro quota is custom and not published.","Each execution is limited to 128 MiB of memory; the default execution limit is 30 seconds (up to 15 minutes on Pro).","Network access is limited to HTTP, and code only sees the inputs, environment variables and files you pass in. Output files are not supported; results come back on stdout or stderr.","Self-hosting is available on the custom-priced Enterprise plan."]},"bestFor":["Running short LLM-generated scripts where per-request pricing is simpler than compute metering","Multi-language snippets in Python, JavaScript, TypeScript, Ruby or PHP","Free hobby projects within 100,000 executions a month"],"watchOut":["128 MiB memory and 30-second executions on the free plan","No shell, persistent state or general Linux environment","Third-party packages only for Python and JavaScript","Jump from free to $250 per month with no published plan in between"],"sources":[{"url":"https://riza.io/pricing","quote":"Up to 100,000 requests per month","checked":"2026-11-13"},{"url":"https://riza.io/pricing","quote":"1,000,000 requests per month included","checked":"2026-11-13"},{"url":"https://riza.io/pricing","quote":"Up to 15-minute execution time limit per request","checked":"2026-11-13"},{"url":"https://riza.io/pricing","quote":"|Data transfer per month |100 GB |500 GB |Custom |","checked":"2026-11-13"},{"url":"https://riza.io/pricing","quote":"A service for executing untrusted Python, JavaScript, TypeScript, Ruby and PHP.","checked":"2026-11-13"},{"url":"https://riza.io/pricing","quote":"Network access is limited to HTTP.","checked":"2026-11-13"},{"url":"https://docs.riza.io/interpreters/limits","quote":"| Memory consumption | 128Mib |","checked":"2026-11-13"},{"url":"https://docs.riza.io/reference/security","quote":"Code is executed inside a secure WebAssembly sandbox.","checked":"2026-11-13"},{"url":"https://docs.riza.io/reference/security","quote":"By default, code sandboxes have no filesystem or network access.","checked":"2026-11-13"},{"url":"https://docs.riza.io/getting-started/client-libraries","quote":"Access the Riza API from Python, Node.js or Go","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/riza","score":{"code-sandbox":{"total":52,"rank":56,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"runloop","name":"Runloop","company":"Runloop","website":"https://runloop.ai","docsUrl":"https://docs.runloop.ai/docs/overview/what-is-runloop","pricingUrl":"https://www.runloop.ai/pricing","categories":["code-sandbox","dev-box"],"tagline":"VM-isolated \"Devbox\" sandboxes for coding agents, with blueprints, snapshots and suspend/resume.","summary":"Runloop runs agent workloads in Devboxes, VM-isolated Linux sandboxes created through Python and TypeScript SDKs or a REST API. Compute is billed per second for CPU and memory while a Devbox is initializing, running, suspending or resuming, with separate per-GB-hour charges for Devbox, blueprint, snapshot and object storage. The Basic plan has no monthly fee; Pro costs $250/month and adds suspend/resume; new accounts get a Pro trial with $50 of usage credit.","isolation":"Virtual machine (microVM) per Devbox","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","REST API"],"regions":"Not published on the pages checked; Enterprise can deploy into the customer's own VPC.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.108,"gibHour":0.0252,"instanceHour":null,"instance":null,"storageGibMonth":0.2499228,"billingIncrement":"per second","billedWhile":"Compute (CPU and memory) while the Devbox status is initializing, running, suspending or resuming; storage keeps billing while suspended.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Basic","monthlyFee":0,"includedUsageUsd":0,"limits":"Free subscription plus usage-based compute; 100 GB free storage; Devboxes, blueprints and snapshots; email support. Suspend/resume is listed as a Pro feature.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":250,"includedUsageUsd":0,"limits":"$250/month plus usage; 1 TB free storage; adds suspend and resume, repo connections, beta features and Slack support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"New accounts start on a Pro trial with a one-time $50 usage credit and no credit card (3 concurrent Devboxes, sizes up to MEDIUM, 1-hour max keep-alive). After that, the Basic plan has no monthly fee and bills usage only.","notes":["Per-second list rates: $0.00003/CPU-s ($0.108/CPU-hour) and $0.000007/GB-s ($0.0252/GB-hour). Runloop labels memory in GB on the pricing page and GiB in the size docs; we use the GB rate as the GiB rate.","Devbox storage is $0.00034236/GB-hour; $/GB-month = 0.00034236 x 730 hours = $0.2499228/GB-month (730-hour month is our assumption). Basic includes 100 GB and Pro 1 TB of free storage; the page does not say exactly which storage types the free allowance covers.","Blueprint, snapshot and object storage cost $0.000072/GB-hour (about $0.0526/GB-month at 730 hours); blueprint builds cost $0.252/hour; active Axons $0.006/axon-hour; inactive Axon storage $0.20/GB-month.","Suspended Devboxes stop CPU and memory billing but keep paying for storage; snapshots and storage objects accrue storage charges until deleted.","Fixed sizes from the docs (CPU + memory + disk at list rates): X_SMALL 0.5 CPU/1 GB/4 GB $0.0806/hr, SMALL 1/2/4 $0.1598/hr, MEDIUM 2/4/8 $0.3195/hr, LARGE 2/8/16 $0.4231/hr, X_LARGE 4/16/16 $0.8407/hr, XX_LARGE 8/32/16 $1.6760/hr. Custom sizes go up to 16 CPU and 64 GiB memory.","Default Devbox lifetime is 1 hour (keep_alive_time_seconds is configurable); no overall maximum lifetime is published for paid plans, so we show no maximum session.","Pricing page says 'No minimums' for usage. The Basic card also says '1st month FREE' without explaining what it applies to.","Enterprise (custom pricing, VPC deployment, custom free storage) is not listed as a plan because no fee is published. Self-hosting is possible only through Enterprise 'Deploy to VPC'.","The docs index describes Devboxes as having 'optional GPU', but no GPU types or prices are published, so GPU support is marked unknown."]},"bestFor":["Coding-agent pipelines that need snapshots, reusable blueprint images and suspend/resume","Long-lived but intermittent agent sessions on Pro, where suspended Devboxes stop compute billing","Teams that need network egress policies and an Enterprise VPC deployment option"],"watchOut":["CPU costs $0.108/CPU-hour and is charged on the allocated size while running, not on active use","Suspend/resume is a Pro feature, and Pro costs $250/month on top of usage","Devboxes default to a 1-hour lifetime; the trial caps keep-alive at 1 hour and sizes at MEDIUM","Suspend keeps only disk state; processes must be restarted after resume"],"sources":[{"url":"https://www.runloop.ai/pricing","quote":"Devbox CPU $0.108 / CPU-hr Devbox Memory $0.0252 / GB-hr","checked":"2026-11-14"},{"url":"https://www.runloop.ai/pricing","quote":"Pro Recommended $250 /month + usage","checked":"2026-11-14"},{"url":"https://www.runloop.ai/pricing","quote":"Suspended devboxes do not accrue compute or memory charges. Billing for CPU and RAM stops the moment suspension begins. Storage charges continue while a devbox is suspended","checked":"2026-11-14"},{"url":"https://docs.runloop.ai/docs/devboxes/configuration/sizes","quote":"Compute is billed only while devbox status is one of [initializing, running, suspending, resuming].","checked":"2026-11-14"},{"url":"https://docs.runloop.ai/docs/devboxes/configuration/sizes","quote":"Min is 0.5 core, max is 16 cores. ... Min is 1GiB, max is 64GiB.","checked":"2026-11-14"},{"url":"https://docs.runloop.ai/docs/devboxes/start-stop","quote":"The default lifetime is 1 hour, but this can be configured when you create the Devbox","checked":"2026-11-14"},{"url":"https://docs.runloop.ai/docs/overview/your-runloop-trial","quote":"It includes $50 in free credits, and no credit card is required to sign up.","checked":"2026-11-14"},{"url":"https://docs.runloop.ai/llms.txt","quote":"only disk state survives suspend (in-memory is lost); background processes must be restarted after resume; suspended devboxes still incur storage charges.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/runloop","score":{"code-sandbox":{"total":62,"rank":36,"of":63},"dev-box":{"total":61,"rank":24,"of":38}},"comparablePrice":{"code-sandbox":55.7568,"dev-box":69.696},"benchmarks":[{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":8.29,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":12,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":24.8,"unit":"s","better":"lower","date":"2026-09-29","rank":11,"of":11}]},{"slug":"runs-on","name":"RunsOn","company":"RunsOn","website":"https://runs-on.com","docsUrl":"https://runs-on.com/docs/","pricingUrl":"https://runs-on.com/pricing/","categories":["ci-runner"],"tagline":"Self-hosted GitHub Actions runners in your own AWS account: a flat yearly license, with EC2 billed by AWS at cost.","summary":"RunsOn is a control plane you deploy into your own AWS account with one CloudFormation or Terraform stack; it launches a fresh EC2 instance for every GitHub Actions job and removes it afterwards. You pay RunsOn a flat yearly license sized by jobs per month (from $350 a year), and AWS bills you directly for the instances, disks, cache bucket and a small control plane. On its pricing page RunsOn estimates a 2 vCPU x64 spot runner at about $0.0009 a minute in us-east-1. It is free for open-source, non-profit and personal non-commercial use.","isolation":"Fresh EC2 instance per job in your own AWS account","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":"https://github.com/runs-on/runs-on","lastRelease":null,"sdks":["GitHub Actions runs-on labels","CloudFormation","Terraform"],"regions":"Any AWS account and region where you deploy the stack; one license covers all your stacks. Example prices are for us-east-1.","features":{"pauseResume":false,"snapshots":null,"gpu":true,"customImages":true,"networkControls":null,"maxSessionHours":null,"maxVcpu":96,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.054,"instance":"AWS m8a.large spot, 2 vCPU x64, with a 30 GB gp3 root volume; paid to AWS, not RunsOn ($0.0009 per minute, us-east-1 7-day average)","storageGibMonth":null,"billingIncrement":"per second (AWS EC2 and EBS billing); license billed yearly","billedWhile":"AWS bills the instance and its volume only while a job runs; the control plane (about $7 a month) runs continuously in your account. The license is a flat yearly fee.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[{"type":"NVIDIA T4 (AWS g4dn.xlarge spot, 4 vCPU; paid to AWS)","perHour":0.234}],"plans":[{"name":"Free (non-commercial)","monthlyFee":0,"includedUsageUsd":0,"limits":"Full product for open-source, non-profit and personal non-commercial projects, in exchange for a public link back to RunsOn. AWS costs still apply.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Starter","monthlyFee":29.166667,"includedUsageUsd":0,"limits":"$350 a year (€300) for under 50,000 jobs a month; all GitHub orgs and AWS accounts of one company; email support. AWS compute billed separately.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Growth","monthlyFee":87.5,"includedUsageUsd":0,"limits":"$1,050 a year (€900) for under 200,000 jobs a month. AWS compute billed separately.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":175,"includedUsageUsd":0,"limits":"$2,100 a year (€1,800) for under 500,000 jobs a month. AWS compute billed separately.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Sponsorship","monthlyFee":350,"includedUsageUsd":0,"limits":"$4,200 a year (€3,600), unlimited jobs, full server and agent source code for review, priority support via Slack Connect. Also sold on AWS Marketplace. AWS compute billed separately.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"15-day free trial of Starter, Growth or Scale (payment details collected at checkout); AWS usage during the trial is billed by AWS. Open-source, non-profit and personal non-commercial projects get a free license.","notes":["There are two bills. The RunsOn license is a flat yearly fee per company (monthly figures here are the yearly price divided by 12). Compute, storage, the S3 cache bucket, data transfer and the control plane (from about $7 a month) are billed by AWS on your own account with no markup from RunsOn.","Rate shown is RunsOn's published estimate of the AWS bill for a 2 vCPU x64 spot runner (m8a.large, 7-day average spot price in the cheapest us-east-1 zone, including a 30 GB gp3 root volume): $0.0009 per minute x 60 = $0.054 per hour. Spot prices change, differ by region and can be interrupted; on-demand instances cost more.","Other examples from the same table: 2 vCPU arm64 (m8g.large) $0.0009 per minute, 2 vCPU Windows (m8i.large) $0.0020 per minute, 4 vCPU T4 GPU (g4dn.xlarge) $0.0039 per minute ($0.234 per hour).","License tiers are set by jobs per month and are not enforced as hard limits; only two consecutive months above a tier move you up, prorated. Prices are shown in EUR by default with USD equivalents.","Runners use AMIs built from GitHub's runner images, refreshed every 15 days, or your own AMI. Instance families and sizes are chosen per job in the runs-on label.","The public repository holds the CloudFormation templates and docs under the MIT license; the server and agent source code is only provided to Sponsorship customers."]},"bestFor":["Companies with heavy GitHub Actions usage that want compute at AWS spot prices","Teams that need runners inside their own AWS network, IAM and VPC","Jobs that need specific EC2 families, GPUs or custom AMIs"],"watchOut":["You operate the stack in your own AWS account and pay AWS separately","Spot prices vary and spot instances can be reclaimed mid-job","The yearly license is due even in months with little CI use","AWS only; no macOS runners"],"sources":[{"url":"https://runs-on.com/pricing/","quote":"AWS bills you for the instances at spot prices, with no markup.","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"| 2cpu | m8a.large | $0.0009 | $0.0060 | 7× | 3,812 | 2,269 | +68% |","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"spot prices (7-day average, cheapest zone) · us-east-1 · incl. 30GB gp3 @ 400 MB/s billed by AWS, to the second","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"Starter < 50k jobs/mo €300 / yr","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"Checkout supports EUR and USD","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"From about $7 a month; no egress to a third party.","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"15 days, with payment details collected at checkout.","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"The full product for non-profit, open-source and personal non-commercial projects, in exchange for a public link back to runs-on.com.","checked":"2026-11-13"},{"url":"https://runs-on.com/pricing/","quote":"| 4cpu·T4 | g4dn.xlarge | $0.0039 | $0.0520 | 13× |","checked":"2026-11-13"},{"url":"https://github.com/runs-on/runs-on","quote":"It launches an isolated EC2 runner for every job with full GitHub Linux runner-image compatibility.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/runs-on","score":{"ci-runner":{"total":75,"rank":2,"of":10}},"comparablePrice":{"ci-runner":5.4},"benchmarks":[]},{"slug":"sailboxes","name":"Sailboxes","company":"Sail Research","website":"https://sailresearch.com/sailboxes","docsUrl":"https://docs.sailresearch.com/sailboxes","pricingUrl":"https://docs.sailresearch.com/pricing","categories":["dev-box"],"tagline":"Persistent Linux VMs with no runtime cap, billed only for the CPU time, RAM and disk actually used.","summary":"Sailboxes are isolated Linux VMs from Sail Research, the inference provider, built for agents that run for hours or days. There is no runtime limit: a Sailbox can sleep or pause with memory and running processes saved, and checkpoints can start new Sailboxes from the same state. Billing follows actual use rather than reserved size: CPU time, physical RAM held and disk used are metered while the Sailbox runs, up to 64 vCPUs and 256 GiB per box. The free plan gives $5 of credit a month once a payment method is attached; Pro costs $250 a month with $100 of included credit.","isolation":"Separate Linux VM per Sailbox, each with its own kernel","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","Rust","CLI","REST API"],"regions":"Not stated","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":64,"maxMemoryGib":256},"pricing":{"rates":{"vcpuHour":0.015,"gibHour":0.008,"instanceHour":null,"instance":null,"storageGibMonth":0.30003,"billingIncrement":"Sampled every second and reported in roughly one-minute batches; not rounded up to whole minutes","billedWhile":"CPU time, RAM held and disk used are billed only while the Sailbox runs; sleeping, paused, checkpointing and cold-starting time is not billed. Volume storage is billed for every hour a volume exists.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":5,"limits":"$5 of credit every month once a payment method is attached; usage paid from prepaid credits; up to 4 seats; 200 reserved vCPUs across active Sailboxes; $0.01 fee per Sailbox created.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":250,"includedUsageUsd":100,"limits":"$100 of included credit every month plus an extra $150 in the first month; 10,000 reserved vCPUs across active Sailboxes; creation fees waived; unlimited seats; priority support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: $5 of credit every month after a payment method is attached; usage beyond it is paid from prepaid credits.","notes":["CPU is billed on CPU time used ($0.015 per vCPU-hour), memory on physical RAM the Sailbox holds including the guest's file cache ($0.008 per GiB-hour), and local NVMe disk on space used ($0.0007 per GiB-hour) while running. Configured limits do not add charges.","The storage rate we show is volume storage: $0.000411 per GiB-hour x 730 hours = $0.30003 per GiB-month, billed for as long as a volume exists, even while attached Sailboxes sleep. Local disk used by a running Sailbox is a separate $0.0007 per GiB-hour (about $0.51 per GiB-month if it ran all month).","Each new Sailbox on the Free plan costs $0.01, including copies started from checkpoints; Pro and Enterprise waive it.","Per-Sailbox limits: 1 to 64 vCPUs, 16 to 256 GiB of memory and 8 to 1,024 GiB of disk, rounded up to the next power of two. Configured vCPUs count toward the organization's reserved-vCPU allowance while the Sailbox is active.","When credits run out, Sail pauses running and sleeping Sailboxes and blocks new spend; it does not terminate them.","Enterprise pricing is custom, billed monthly in arrears."]},"bestFor":["Agents and jobs that run for days without a runtime cap","Workloads with spiky or unknown memory needs, since only RAM actually held is billed","Long tasks that need to pause and later resume with processes and memory intact"],"watchOut":["Free plan credit needs a payment method attached, and each new Sailbox costs $0.01 on Free","Billed memory includes the Linux file cache inside the Sailbox, so an idle running box still accrues charges","Regions and GPU support are not documented on the pages checked","Pro costs $250 a month, a large step up from the free plan"],"sources":[{"url":"https://docs.sailresearch.com/pricing","quote":"Used vCPU/hour (CPU time of the Sailbox's virtual CPUs) | $0.015","checked":"2026-11-13"},{"url":"https://docs.sailresearch.com/pricing","quote":"Used RAM (GiB)/hour (Physical RAM the Sailbox holds on the host, including its Linux filesystem cache) | $0.008","checked":"2026-11-13"},{"url":"https://docs.sailresearch.com/pricing","quote":"Volume storage (GiB)/hour | $0.000411","checked":"2026-11-13"},{"url":"https://docs.sailresearch.com/pricing","quote":"$100/month in included credits, with an additional $150 in your first month","checked":"2026-11-13"},{"url":"https://docs.sailresearch.com/pricing","quote":"$5/month in free credits when you attach a payment method","checked":"2026-11-13"},{"url":"https://docs.sailresearch.com/sailboxes-pricing.md","quote":"Running usage is sampled every second and reported in roughly one-minute batches. Usage is not rounded up to whole minutes.","checked":"2026-11-13"},{"url":"https://docs.sailresearch.com/sailboxes-pricing.md","quote":"Memory: 16–256 GiB per Sailbox. Requests above 256 GiB are rejected.","checked":"2026-11-13"},{"url":"https://sailresearch.com/sailboxes","quote":"Run agent workloads in separate Linux VMs, each with its own kernel and isolated environment.","checked":"2026-11-13"},{"url":"https://sailresearch.com/sailboxes","quote":"Pause for a minute or a month. Resume with memory and processes intact","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/sailboxes","score":{"dev-box":{"total":86,"rank":1,"of":38}},"comparablePrice":{"dev-box":8.36},"benchmarks":[]},{"slug":"sandbase","name":"SandBase","company":"AXIOMCLOUD PTE. LTD.","website":"https://www.sandbase.ai","docsUrl":"https://www.sandbase.ai/docs/sandbox/","pricingUrl":"https://www.sandbase.ai/pricing","categories":["code-sandbox"],"tagline":"E2B-compatible agent sandboxes, free for now, on the same API key as a catalog of AI models and data APIs.","summary":"SandBase, run by Singapore-based AXIOMCLOUD, sells AI models, data APIs and managed agents on one prepaid balance, and adds an agent sandbox that speaks the E2B protocol. You keep the official E2B JavaScript or Python SDK and point it at SandBase's endpoint with a SandBase key, then create an instance from a template, run commands, read and write files, reach ports inside it, and pause and reconnect where the template supports it. SandBase describes the sandbox as a gateway: the runtime and isolation come from the underlying provider behind each template, which it does not name. The public templates are 2 vCPU with 2 GB of memory. Sandboxes are free for now, with possible fair-use caps.","isolation":"Not stated; SandBase is a gateway and says isolation depends on the underlying provider and template","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["E2B Python SDK","E2B JavaScript SDK","REST API"],"regions":"Not stated.","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: sandboxes are currently free","billedWhile":"Not billed for now; creating, running and stopping sandboxes does not use the account balance","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Start","monthlyFee":0,"includedUsageUsd":0,"limits":"Starter credits for models and APIs; sandboxes are free for now and fair-use caps may apply. The credit amount is not stated.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Sandboxes are free for now, with possible fair-use caps. SandBase says it will announce pricing before this changes.","notes":["No sandbox compute rates are published; the free offer is described as temporary and labelled a limited-time offer in the quickstart.","The four public templates listed in SandBase's own repository are each 2 vCPU, 2048 MB of memory and about 23 GB of disk. Organizations may also see private templates, whose sizes are not published, so no maximum size is stated.","Pause and resume, snapshots, template builds and network controls depend on the provider behind each template; SandBase tells users to check support per template.","Model and data API calls are billed separately from the same prepaid balance."]},"bestFor":["Trying an E2B-style sandbox at no cost by changing only the key, API URL and domain","Builders who already buy models or data APIs from SandBase and want code execution on the same key","Short-lived agent tasks that run commands and exchange files"],"watchOut":["The isolation technology and the underlying provider are not disclosed, so you cannot judge the security boundary from SandBase's pages","The free offer is temporary and future sandbox pricing is unknown","Upstream E2B templates may not be available; you must pick from the templates visible to your organization"],"sources":[{"url":"https://www.sandbase.ai/pricing","quote":"Agent sandboxes are free for now. Fair-use caps may apply, and pricing will be announced before it changes.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/pricing","quote":"Starter credits to try everything.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/sandbox","quote":"Sandboxes are free for now: creating, running and stopping them does not touch your SandBase balance.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/sandbox","quote":"The runtime and isolation technology behind the selected provider and template.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/sandbox","quote":"Isolation, network controls, and persistence depend on the selected provider and template.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/docs/sandbox/quickstart","quote":"Use the official E2B SDK with your SandBase API key to create an isolated sandbox, run a command, and read and write files.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/docs/sandbox/quickstart","quote":"Pause/resume must be supported by the selected template and provider.","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/docs/sandbox/quickstart","quote":"Free for a limited time","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/docs/sandbox/quickstart","quote":"Template builds, snapshots, events, webhooks, and advanced networking need separate capability checks.","checked":"2026-11-15"},{"url":"https://raw.githubusercontent.com/sandbaseai/sandbase/main/catalog/sandbox.json","quote":"\"memory_mb\": 2048","checked":"2026-11-15"},{"url":"https://raw.githubusercontent.com/sandbaseai/sandbase/main/catalog/sandbox.json","quote":"\"cpu_count\": 2","checked":"2026-11-15"},{"url":"https://www.sandbase.ai/terms","quote":"offered by AXIOMCLOUD PTE. LTD., a company registered in Singapore","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-15","page":"https://aisandboxhunt.com/sandbase","score":{"code-sandbox":{"total":49,"rank":57,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"sandbox0","name":"Sandbox0","company":"Sandbox0","website":"https://sandbox0.ai","docsUrl":"https://sandbox0.ai/docs/sandbox","pricingUrl":"https://sandbox0.ai/pricing","categories":["dev-box"],"tagline":"gVisor sandboxes that keep the whole root filesystem across pauses, billed per second on memory size only.","summary":"Sandbox0 gives each user's agent a persistent cloud computer in a gVisor sandbox whose entire writable root filesystem is kept across pauses and grows automatically as it fills. Compute pauses between tasks and the same sandbox resumes later; by default processes restart fresh, with memory checkpoints available as an experimental option. Compute is billed by the second at $0.015 per GiB-hour of the configured memory limit, with no separate CPU charge, and retained changes cost $0.02 per GiB-month. Network traffic is currently free.","isolation":"gVisor (runsc) sandbox per user","isolationKind":"gvisor","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","JavaScript/TypeScript","Go","CLI","REST API"],"regions":"Not stated; quotas and templates are shown per region in the dashboard.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0,"gibHour":0.015,"instanceHour":null,"instance":null,"storageGibMonth":0.02,"billingIncrement":"Per second (about $0.000004167 per GiB-second), no one-hour minimum","billedWhile":"Compute while the sandbox runs, including idle running time, on its configured memory limit. Paused sandboxes pay only retained storage.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":0,"limits":"Prepaid funds, no plan fee. Default quotas: 20 running sandboxes per team, unlimited paused sandboxes, 5 sandbox claims per second, 100 API requests per second; increases on request.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"One-time $5 usage credit for eligible new users after a credit card is saved; adding the card charges nothing.","notes":["There is no separate CPU charge: compute is priced only on the configured memory limit, and CPU is derived from memory by the platform. We show the CPU rate as $0.","Storage is $0.02 per GiB-month on changed root filesystem blocks and retained memory checkpoints, prorated over 730 hours. Shared base images, duplicate snapshot or fork references and platform recovery copies are not billed.","Network ingress, egress (free for now) and object-store requests currently cost nothing.","Default pause keeps files and starts fresh processes; memory checkpoints that restore process state are experimental.","Self-hosting is no longer supported and bring-your-own-cloud is planned but not available."]},"bestFor":["Per-user agent computers where installed tools and system configuration must persist","Bursty workloads that pause between tasks, since paused sandboxes pay only for stored changes","CPU-heavy work at a small memory size, since only memory is metered"],"watchOut":["gVisor is a user-space kernel, a weaker boundary than a hardware VM","Running processes restart after a default pause; memory checkpoints are experimental","Maximum CPU and memory per sandbox and regions are not published","Egress is free only for now"],"sources":[{"url":"https://sandbox0.ai/pricing","quote":"Compute | $0.015 per GiB-hour | Billed by the second at ≈$0.000004167/GiB-second. Rate follows the running memory limit.","checked":"2026-11-13"},{"url":"https://sandbox0.ai/pricing","quote":"Retained storage | $0.02 per GiB-month","checked":"2026-11-13"},{"url":"https://sandbox0.ai/pricing","quote":"No separate CPU charge; CPU limits still apply.","checked":"2026-11-13"},{"url":"https://sandbox0.ai/pricing","quote":"Eligible new users receive a one-time $5 usage credit after Stripe confirms a saved credit card.","checked":"2026-11-13"},{"url":"https://sandbox0.ai/pricing","quote":"Active sandboxes | 20 running at once | Per team","checked":"2026-11-13"},{"url":"https://sandbox0.ai/","quote":"alloc/sb-agent-7f29 gVisor","checked":"2026-11-13"},{"url":"https://sandbox0.ai/docs/sandbox.md","quote":"Default filesystem-only pause does not preserve running processes, memory, sockets, PID state, or live REPL sessions.","checked":"2026-11-13"},{"url":"https://sandbox0.ai/docs/sandbox.md","quote":"prove network, RootFS, runsc, and procd command readiness for that generation","checked":"2026-11-13"},{"url":"https://sandbox0.ai/llms.txt","quote":"Sandbox0 Cloud is the supported service. Self-hosting is no longer supported.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/sandbox0","score":{"dev-box":{"total":67,"rank":14,"of":38}},"comparablePrice":{"dev-box":13.2},"benchmarks":[]},{"slug":"simular-box","name":"Box by Simular","company":"Simular","website":"https://box.simular.ai","docsUrl":"https://box.simular.ai/docs","pricingUrl":"https://www.simular.ai/pricing","categories":["dev-box","browser"],"tagline":"A free, persistent cloud Ubuntu desktop with Chrome and a shell that Claude, ChatGPT and coding agents drive over MCP.","summary":"Box by Simular gives an AI agent a cloud Linux computer of its own: an Ubuntu desktop with Chrome, files and a persistent shell, driven through one hosted MCP server by Claude, ChatGPT, Gemini, Claude Code, Codex, Cursor or any MCP client. Each account gets a box that sleeps when idle, wakes in about a second with files and logins kept, and can be watched live and taken over in the browser. API keys can be stored so commands on the box use them while the agent only sees a stand-in. The box itself is free; the computer-use model calls it makes draw on the Simular account's credits. A box left asleep for 14 days in a row is deleted.","isolation":"Not stated; a cloud Ubuntu desktop per account","isolationKind":"unknown","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["MCP server","CLI"],"regions":"Not stated","features":{"pauseResume":true,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not stated; the box itself is free","billedWhile":"The box has no charge of its own; the model calls it makes use Simular account credits.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"One cloud Linux desktop per account at no charge; model calls use the account's credits, which can be topped up or raised with a Simular plan.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"The box itself is free. Simular's free account plan refreshes credits daily for model calls; the daily amount is not stated.","notes":["No hourly, CPU, memory or disk price is published for the box, and no machine size is stated.","How many credits a task on the box uses is not stated. Simular's general pricing page lists a $50 a month plan with 50,000 credits ($50 of usage) for AI models and computer time, but that page describes its Sai assistant and does not mention the box.","Simular's guide says the box is free during its beta and will have a freemium version afterwards.","A box that stays asleep for 14 days in a row is deleted with its files; any wake resets the count."]},"bestFor":["Giving Claude, ChatGPT or a coding agent a real desktop and browser through a single remote MCP server","One shared computer that several agents use, with logins and files kept between tasks","Watching an agent's screen live and stepping in for sign-ins, 2FA codes or payments"],"watchOut":["Use is through the MCP server only; there is no SDK or REST API for creating machines from code","Machine size, isolation and regions are not documented","Model credits per task are not published, and pricing may change after the beta","An unused box is deleted after 14 days asleep"],"sources":[{"url":"https://box.simular.ai/llms.txt","quote":"Cost: the box itself is free. Its model calls use your account's credits","checked":"2026-11-15"},{"url":"https://box.simular.ai/llms.txt","quote":"A box asleep for 14 days in a row is deleted, with its files and saves; any wake resets the count.","checked":"2026-11-15"},{"url":"https://box.simular.ai/docs/mcp.md","quote":"It wakes in about a second when your AI connects; files and state persist while it sleeps.","checked":"2026-11-15"},{"url":"https://box.simular.ai/docs/mcp.md","quote":"One box per account, with Chrome and a persistent shell.","checked":"2026-11-15"},{"url":"https://box.simular.ai/docs","quote":"a cloud Ubuntu desktop with Chrome and a shell, which your AI drives over MCP","checked":"2026-11-15"},{"url":"https://box.simular.ai/docs/secrets.md","quote":"A secret is an API key that commands on your box can use while your AI never sees it.","checked":"2026-11-15"},{"url":"https://www.simular.ai/pricing","quote":"Free · credits refreshed daily","checked":"2026-11-15"},{"url":"https://www.simular.ai/pricing","quote":"50,000 credits a month ($50 of usage) for AI models and computer time","checked":"2026-11-15"},{"url":"https://www.simular.ai/articles/linux-desktop-in-the-cloud","quote":"box by Simular is free during its beta and has a freemium version afterwards.","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-15","page":"https://aisandboxhunt.com/simular-box","score":{"dev-box":{"total":46,"rank":34,"of":38},"browser":{"total":45,"rank":7,"of":12}},"comparablePrice":{"dev-box":null,"browser":null},"benchmarks":[]},{"slug":"solo-agent-substrate","name":"Solo Enterprise for Agent Substrate","company":"Solo.io","website":"https://www.solo.io/products/kagent","docsUrl":"https://docs.solo.io/kagent/latest/","pricingUrl":"https://www.solo.io/pricing","categories":["code-sandbox"],"tagline":"Supported Agent Substrate on your Kubernetes: each agent session in a gVisor or microVM sandbox, suspended when idle.","summary":"Solo.io's enterprise release of Agent Substrate and kagent, generally available since November 2026, runs AI agents on a Kubernetes cluster you operate. Agent Substrate is an open-source runtime, co-maintained by Solo.io and Google, that runs every agent session as an actor in its own gVisor or microVM sandbox on a pool of pre-started worker pods. When a turn ends it snapshots the actor's files and memory to object storage and frees the worker, then restores the actor on any free worker when the next request arrives. kagent adds declarative agents, harnesses for Claude Code, Codex and custom images, short-lived standalone sandboxes for scripts, and an egress gateway that injects credentials so the sandbox holds none. The enterprise build adds single sign-on, access control, tracing, air-gapped installs and support; it needs license keys and is priced by quote.","isolation":"gVisor sandbox per actor by default, or a microVM under Cloud Hypervisor with a Kata Containers kernel; many sandboxed actors share one worker pod, and plain containers are not supported","isolationKind":"gvisor","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/agent-substrate/substrate","lastRelease":"2026-11-05","sdks":["kagent CLI","Kubernetes API (kubectl/YAML)","MCP server","A2A"],"regions":"Any Kubernetes cluster you run: 1.34 or later in certificate compatibility mode, 1.37 or later in the default mode, including OpenShift. Air-gapped installs are supported.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Not applicable: enterprise license by quote; no usage rates are published","billedWhile":"Not applicable: software runs on your own cluster, and you pay your provider for its nodes and storage","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[],"freeTierKind":"recurring","freeTier":"The open-source kagent and Agent Substrate projects are free to run yourself. The enterprise release, with support, single sign-on, access control, tracing and air-gapped installs, needs license keys priced by quote from Solo.io sales.","notes":["No public prices. Solo.io says enterprise pricing depends on scope, environments and support needs.","The enterprise install needs two license keys: one for kagent and one for agentgateway, which runs the router and egress gateway in front of the sandboxes.","Each worker can hold up to 1,000 actors. Each agent actor defaults to a limit of 1 CPU and 1 GiB of memory, which the operator can change; no upper size limit is published.","Standalone sandboxes expire after a set lifetime of at most 24 hours, counted from creation. Agent conversations instead suspend between turns and are cleaned up after seven idle days by default.","A default install requests 2.92 vCPU and 5.25 GiB for the control components before any agent runs. You also provide PostgreSQL for production.","GPU support for the 1.0 release is not stated. Solo.io's Agent Substrate overview page, written before 1.0, says GPUs are temporarily unsupported."]},"bestFor":["Platform teams that want many mostly idle agent sessions on the Kubernetes clusters they already run","Regulated or air-gapped environments that need a supported, self-hosted agent runtime","Running untrusted agent code without giving the sandbox API keys, with default-deny outbound traffic"],"watchOut":["Self-hosted only: you operate the cluster, PostgreSQL, object storage and upgrades","Enterprise pricing is by quote only, and the install needs separate license keys","Recent Kubernetes is required (1.34 or later), and there is no in-place upgrade from the older 0.5.x kagent line"],"sources":[{"url":"https://www.solo.io/press-releases/announcing-kagent-1-0-ga","quote":"today announced the general availability of its enterprise releases of Agent Substrate and kagent","checked":"2026-11-15"},{"url":"https://www.solo.io/press-releases/announcing-kagent-1-0-ga","quote":"Enterprise support and deployment options, including air-gapped installations","checked":"2026-11-15"},{"url":"https://www.solo.io/blog/announcing-kagent-1-0-ga","quote":"Every actor runs in its own gVisor or microVM sandbox (Substrate doesn't support plain containers).","checked":"2026-11-15"},{"url":"https://www.solo.io/blog/announcing-kagent-1-0-ga","quote":"Substrate snapshots its workspace (and optionally its entire memory) and gives the worker back.","checked":"2026-11-15"},{"url":"https://www.solo.io/blog/announcing-kagent-1-0-ga","quote":"idle conversations clean themselves up (after seven days by default)","checked":"2026-11-15"},{"url":"https://www.solo.io/blog/announcing-kagent-1-0-ga","quote":"an enterprise-ready distribution of Agent Substrate for running AI agents and Sandboxes on any distribution of Kubernetes (including OpenShift)","checked":"2026-11-15"},{"url":"https://www.solo.io/pricing","quote":"Yes. The open-source projects agentgateway, kgateway, kagent and agentregistry are free to use.","checked":"2026-11-15"},{"url":"https://www.solo.io/pricing","quote":"Enterprise pricing depends on your scope, environments and support needs.","checked":"2026-11-15"},{"url":"https://www.solo.io/pricing","quote":"Contact sales and the Solo.io team will put together pricing for your needs.","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/architecture/agent-substrate/","quote":"The gVisor and micro-VM Workers that Agent Substrate ships both report a capacity of 1000 Actors.","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/architecture/agent-substrate/","quote":"default to 1 CPU and 1Gi of memory","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/architecture/agent-substrate/","quote":"or a micro-VM that runs the workload under Cloud Hypervisor with a Kata Containers kernel and root image","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/reference/changelog/release-notes/","quote":"Every agent runs in a gVisor sandbox by default.","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/system-requirements/","quote":"such as a managed distribution that pins a Kubernetes version earlier than 1.37, choose compatibility mode","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/system-requirements/","quote":"2.92 virtual CPUs (vCPU) and 5.25 GiB","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/substrate-runtime/sandboxing/","quote":"The egress policy is default-deny.","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/about/substrate-runtime/standalone-sandboxes/","quote":"which the controller caps at 86400s","checked":"2026-11-15"},{"url":"https://docs.solo.io/kagent/latest/setup/licensing/","quote":"The Agent Substrate chart takes a separate Solo Enterprise for agentgateway license key","checked":"2026-11-15"},{"url":"https://www.solo.io/discover/agent-substrate","quote":"GPUs are temporarily unsupported","checked":"2026-11-15"},{"url":"https://github.com/agent-substrate/substrate/releases/tag/v1.0.0","quote":"v1.0.0","checked":"2026-11-15"}],"lastChecked":"2026-11-15","added":"2026-11-15","page":"https://aisandboxhunt.com/solo-agent-substrate","score":{"code-sandbox":{"total":68,"rank":25,"of":63}},"comparablePrice":{"code-sandbox":null},"benchmarks":[]},{"slug":"steel","name":"Steel","company":"Nen Labs, Inc.","website":"https://steel.dev","docsUrl":"https://docs.steel.dev","pricingUrl":"https://steel.dev/pricing","categories":["browser"],"tagline":"Cloud browser API for agents with pay-per-hour sessions and an open-source browser server you can self-host.","summary":"Steel runs cloud browser sessions that agents control over CDP with Puppeteer, Playwright or Selenium, or through its own SDKs, CLI and MCP server. Browser time is billed by the minute at $0.10 per hour on the free Launch plan and $0.08 on the $250/month Scale plan, with proxy bandwidth and CAPTCHA solves charged separately. New accounts get $30 of one-time credit. The underlying browser server, Steel Browser, is Apache-2.0 licensed and can be run with Docker.","isolation":"Each session is an isolated cloud browser with its own filesystem; the docs mention a session VM but do not name the technology","isolationKind":"unknown","openSource":true,"selfHost":true,"license":"Apache-2.0","repoUrl":"https://github.com/steel-dev/steel-browser","lastRelease":"2026-10-31","sdks":["TypeScript/Node.js","Python","CLI","REST API","MCP server"],"regions":"Hosted browsers run in one region, us-east; proxy geolocation sets the public IP location.","features":{"pauseResume":null,"snapshots":null,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":1,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"Per minute, rounded up","billedWhile":"While a browser session is live, until it is released or times out","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":0.08},"gpu":[],"plans":[{"name":"Launch","monthlyFee":0,"includedUsageUsd":0,"limits":"$0.10/browser-hour, $10/GB proxy, $3 per 1,000 CAPTCHA solves; 10 concurrent sessions; sessions up to 15 minutes; 60 requests per minute; up to 3 seats.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Scale","monthlyFee":250,"includedUsageUsd":100,"limits":"$100 of usage credit every month; $0.08/browser-hour, $6/GB proxy, $1 per 1,000 CAPTCHA solves; 100 concurrent sessions; sessions up to 1 hour; 600 requests per minute; dedicated IPs at $5 each.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"$30 of one-time credit on the Launch plan, no credit card needed.","notes":["The per-hour price shown is the Scale plan rate ($0.08/browser-hour, with a $250 monthly fee that includes $100 of usage credit). On the free Launch plan the rate is $0.10/browser-hour.","Proxy bandwidth is $10/GB on Launch and $6/GB on Scale. CAPTCHA solves are $3 per 1,000 on Launch and $1 per 1,000 on Scale. The /scrape browser tool is $5 per 1,000 calls.","Sessions last 5 minutes by default and can be set up to the plan maximum (15 minutes on Launch, 1 hour on Scale, up to 24 hours on Enterprise). The timeout cannot be changed on a live session.","Stealth browser and reserved browser pools are Enterprise-only; Enterprise pricing is custom.","Steel Browser, the open-source server behind the hosted service, is Apache-2.0 and can be self-hosted with Docker at no license cost."]},"bestFor":["Short agent or scraping sessions billed by the minute without a monthly fee","Teams that want the option to run the same browser server themselves","Agents driven from a CLI or MCP client as well as from code"],"watchOut":["Sessions on the free plan end after 15 minutes; 1 hour is the self-serve maximum","The $30 starting credit is one-time, not monthly","Hosted browsers run only in us-east","Stealth browser is only on Enterprise"],"sources":[{"url":"https://steel.dev/pricing","quote":"Start for free with $30 of credits, and pay only for what your agents run.","checked":"2026-11-13"},{"url":"https://steel.dev/pricing","quote":"Included usage:$100 of credits every month","checked":"2026-11-13"},{"url":"https://steel.dev/pricing","quote":"CAPTCHA solves |$3/1k |$1/1k |Custom","checked":"2026-11-13"},{"url":"https://steel.dev/pricing","quote":"Max session time |15 minutes |1 hour |Up to 24 hours","checked":"2026-11-13"},{"url":"https://docs.steel.dev/overview/pricinglimits","quote":"Browser hours are billed by the minute, rounded up. Launch is $0.10/hour, Scale is $0.08/hour","checked":"2026-11-13"},{"url":"https://docs.steel.dev/overview/pricinglimits","quote":"Proxy bandwidth is $10/GB on Launch and $6/GB on Scale","checked":"2026-11-13"},{"url":"https://docs.steel.dev/overview/sessions-api/configuration","quote":"Steel runs browsers in one region","checked":"2026-11-13"},{"url":"https://github.com/steel-dev/steel-browser","quote":"Apache-2.0 license","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/steel","score":{"browser":{"total":41,"rank":10,"of":12}},"comparablePrice":{"browser":0.08},"benchmarks":[]},{"slug":"superserve","name":"Superserve","company":"Superserve","website":"https://superserve.ai","docsUrl":"https://docs.superserve.ai/introduction","pricingUrl":"https://www.superserve.ai/pricing/","categories":["code-sandbox","dev-box"],"tagline":"Firecracker microVMs for agents that pause with memory and disk kept, billed per second for vCPU and memory.","summary":"Superserve runs each sandbox in a Firecracker microVM started from a template, driven by TypeScript and Python SDKs, a CLI, a REST API or an MCP server. Compute ($0.0504 per vCPU-hour) and allocated memory ($0.0162 per GiB-hour) bill per second while a sandbox is active; pausing checkpoints memory, processes and files and stops compute billing, and storage bills for retained data. Runtime is unlimited and a paused sandbox is kept until deleted unless an auto-delete window is set. Paid plans are monthly fees that include the same amount of usage.","isolation":"Firecracker microVM per sandbox","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","Python","CLI","REST API","MCP server"],"regions":"Not stated on the pages checked.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":0.07884,"billingIncrement":"per second","billedWhile":"Compute and allocated memory while the sandbox is active; nothing for compute while paused. Storage (shared template once plus each sandbox's retained changes) bills for as long as it is kept, including paused sandboxes.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Developer","monthlyFee":20,"includedUsageUsd":20,"limits":"Includes $20 of usage. Up to 100 concurrent sandboxes, unlimited runtime, up to 4 vCPU and 16 GiB memory per sandbox. Auto Reload adds $5 when the balance drops under $2. Email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":250,"includedUsageUsd":250,"limits":"Includes $250 of usage; extra usage billed at the end of each billing period. Up to 2,000 concurrent sandboxes, unlimited runtime, up to 16 vCPU and 64 GiB memory per sandbox. Priority Slack and email support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Start free with no card: a new account gets a limited free trial for its first sandbox (the credit amount is not stated). Trial sandboxes can use up to 2 vCPU and 4 GiB memory. After the trial credit is used, a card and a plan are required.","notes":["Storage is published as $0.000108 per GiB-hour; the monthly figure we show is that rate times 730 hours ($0.07884 per GiB-month). Storage counts a shared template once plus each sandbox's retained changes, not the full allocated disk.","The same usage rates apply on every plan; included usage on each plan is spent at these rates.","Sandbox vCPU, memory and disk come from the template and cannot be overridden per sandbox. Template defaults are 1 vCPU and 1 GiB; maximums are 2 vCPU / 4 GiB on the free trial, 4 vCPU / 16 GiB on Developer and 16 vCPU / 64 GiB on Pro. The maximum we show is the Pro limit.","Pause checkpoints memory, processes and filesystem and stops compute billing. An optional active-time timeout and an optional idle timeout (60 seconds to 7 days) can pause a sandbox automatically; both are off by default. Paused sandboxes are kept until deleted unless an auto-delete window (max 30 days) is set.","The pricing page's estimates exclude taxes, network and external model API charges; no network or egress rate is published.","Enterprise is custom-priced and offers deployment in your own VPC on AWS, GCP or Azure for early partners; we do not count this as self-hosting. The TypeScript and Python SDKs are open source under Apache-2.0; the platform itself is not described as open source.","No GPU support is described on the pages checked."]},"bestFor":["Long-running agent sessions that pause and resume with memory, processes and files intact.","Branching work: snapshot a sandbox and fork copies to try several approaches in parallel.","Agents that need API credentials without the key entering the sandbox, plus egress allow and deny rules."],"watchOut":["vCPU and allocated memory bill for the whole time a sandbox is active, busy or idle; pause it to stop compute charges.","Sandbox size is fixed by its template; a different size needs a different template.","The free trial amount is not published, and paid use needs a monthly plan.","No region choice or GPU support is documented on the pages checked."],"sources":[{"url":"https://www.superserve.ai/pricing/","quote":"Compute: $0.0504/vCPU-hour · Memory: $0.0162/GiB-hour · Storage: $0.000108/GiB-hour retained.","checked":"2026-11-14"},{"url":"https://www.superserve.ai/pricing/","quote":"Includes $20 of usage","checked":"2026-11-14"},{"url":"https://www.superserve.ai/pricing/","quote":"Up to 16 vCPU and 64 GiB memory per sandbox","checked":"2026-11-14"},{"url":"https://www.superserve.ai/pricing/","quote":"Extra usage billed at the end of each billing period","checked":"2026-11-14"},{"url":"https://www.superserve.ai/pricing/","quote":"Start free, no card required. Then pay for usage.","checked":"2026-11-14"},{"url":"https://docs.superserve.ai/introduction","quote":"A new account gets a limited free trial for its first sandbox.","checked":"2026-11-14"},{"url":"https://docs.superserve.ai/introduction","quote":"Superserve provides sandbox infrastructure to run code in isolated cloud environments powered by Firecracker MicroVMs.","checked":"2026-11-14"},{"url":"https://docs.superserve.ai/sandbox/lifecycle","quote":"By default a paused sandbox is kept forever.","checked":"2026-11-14"},{"url":"https://docs.superserve.ai/sandbox/lifecycle","quote":"`pause()` checkpoints the full VM state (memory, processes, filesystem) to disk and stops billing for compute.","checked":"2026-11-14"},{"url":"https://docs.superserve.ai/templates/build-spec","quote":"Templates specify the VM shape; sandboxes inherit these and cannot override them per-sandbox.","checked":"2026-11-14"},{"url":"https://docs.superserve.ai/sandbox/create","quote":"Egress allow/deny rules.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/superserve","score":{"code-sandbox":{"total":69,"rank":19,"of":63},"dev-box":{"total":69,"rank":11,"of":38}},"comparablePrice":{"code-sandbox":29.1456,"dev-box":36.432},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":377,"unit":"ms","better":"lower","date":"2026-11-13","rank":8,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":57422,"unit":"ms","better":"lower","date":"2026-11-13","rank":9,"of":25},{"source":"aisandboxbench","metric":"Cold start (median)","value":0.52,"unit":"s","better":"lower","date":"2026-10-23","rank":1,"of":9},{"source":"aisandboxbench","metric":"Agent tool-call replay (median)","value":13.147,"unit":"s","better":"lower","date":"2026-10-23","rank":2,"of":9}]},{"slug":"tembo","name":"Tembo","company":"Tembo","website":"https://www.tembo.io","docsUrl":"https://docs.tembo.io","pricingUrl":"https://www.tembo.io/pricing","categories":["dev-box","coding-agent"],"tagline":"Agent orchestration platform that runs each coding session in its own Linux VM, billed per vCPU and GiB.","summary":"Tembo runs coding agents such as Claude Code, Codex, Cursor, Amp and OpenCode on tasks from GitHub, Linear, Slack, Sentry and other tools, with each session in a dedicated Linux VM. VM compute is metered per second at $0.0403 per vCPU-hour plus $0.0130 per GiB-hour while the VM runs, and draws from a monthly dollar allowance that also covers model calls through Tembo's own gateway. Paid plans include more usage than their fee and allow optional overage billing with a spend limit. Sandboxes are set up per session and torn down when it ends, apart from cached project environments.","isolation":"Dedicated Linux VM per session, with nested virtualization for Docker","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["TypeScript","REST API","MCP"],"regions":"Not stated on the pages checked; self-hosted and VPC deployment are offered for organizations.","features":{"pauseResume":null,"snapshots":true,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":32},"pricing":{"rates":{"vcpuHour":0.0403,"gibHour":0.013,"instanceHour":null,"instance":null,"storageGibMonth":null,"billingIncrement":"per second","billedWhile":"Only while the session VM is running. Stopped, snapshotted and otherwise non-running VMs accrue no compute.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Pro","monthlyFee":60,"includedUsageUsd":90,"limits":"$90 of usage a month shared between VM compute and Tembo Gateway model usage; up to 5 users; VM size up to Medium (4 vCPU / 8 GiB); optional overage billing with a spend limit.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Max","monthlyFee":200,"includedUsageUsd":400,"limits":"$400 of usage a month shared between VM compute and Tembo Gateway model usage; up to 10 users; VM size up to XL (16 vCPU / 32 GiB); optional overage billing with a spend limit.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"7-day free trial of the Max plan with an included usage allowance (amount not stated), up to 10 users and VM size up to XL. Sessions stop when the trial or its allowance ends; no overage billing during the trial.","notes":["VM rates: $0.0403 per vCPU-hour + $0.0130 per GiB RAM-hour, metered by the second. Published sizes: Nano 2 vCPU / 2 GiB $0.1066/hour, Micro 2 / 4 $0.1326, Medium 4 / 8 $0.2652, Large 8 / 16 $0.5304, XL 16 / 32 $1.0608, XXL 16 / 64 $1.4768, Ultra 32 / 128 $2.9536.","Self-serve maximum is XL (16 vCPU / 32 GiB) on Max and the trial, and Medium on Pro. XXL and Ultra are listed with prices but are available on request only.","The included allowance is in dollars and is shared with model usage through Tembo Gateway; using your own API key or a Claude, ChatGPT/Codex or SuperGrok subscription costs nothing through Tembo, but VM compute still draws from the allowance. Included allowance does not roll over; prepaid balance does not expire.","Overage is optional on Pro and Max: with it off, new sessions are blocked once the allowance and prepaid balance are used up.","Disk per size: 50 GB (Nano to Medium), 100 GB (Large), 200 GB (XL, XXL and Ultra). No separate storage price is published.","Conflict between vendor pages: the sandbox docs say sandboxes are ephemeral and destroyed when the session ends, while the pricing page lists pausable and resumable sessions and says snapshotted VMs accrue no compute, and the site overview mentions reusable snapshots and on-demand VM keep-alive. Whether a session VM can be paused and resumed later is not clear, so we leave that unknown.","Tembo is mainly an agent orchestration product: VMs are used through Tembo sessions rather than rented on their own. Enterprise pricing (custom allowances and sizes, self-hosting, SSO) is by contact."]},"bestFor":["Teams that want agents triggered from issue trackers, alerts and chat, with each session in its own VM.","Comparing or switching between several coding agents while paying one per-second VM rate.","Workloads that need Docker inside the sandbox, since VMs support nested virtualization."],"watchOut":["VMs are only available inside Tembo sessions; there is no standalone sandbox product.","Pro limits VMs to 4 vCPU / 8 GiB; larger sizes need Max.","The monthly allowance is shared with model usage if you use Tembo's gateway instead of your own key or subscription.","Sandboxes are torn down at the end of a session, apart from cached project environments."],"sources":[{"url":"https://www.tembo.io/pricing","quote":"$0.0403 / vCPU-hour","checked":"2026-11-14"},{"url":"https://www.tembo.io/pricing","quote":"$0.0130 / GiB RAM-hour","checked":"2026-11-14"},{"url":"https://www.tembo.io/pricing","quote":"Metered by the second at the equivalent rate. Stopped and snapshotted VMs do not accrue compute charges.","checked":"2026-11-14"},{"url":"https://www.tembo.io/pricing","quote":"Includes $90 in usage every month","checked":"2026-11-14"},{"url":"https://www.tembo.io/pricing","quote":"Includes $400 in usage every month","checked":"2026-11-14"},{"url":"https://www.tembo.io/pricing","quote":"Start with a 7-day free trial of Max.","checked":"2026-11-14"},{"url":"https://docs.tembo.io/resources/pricing","quote":"Pro ($60/month): $90 monthly usage allowance, up to 5 users, sandbox size up to medium, with overage billing","checked":"2026-11-14"},{"url":"https://docs.tembo.io/resources/pricing","quote":"Tembo measures each running VM in seconds and bills its actual running time at the per-second equivalent of the hourly rate.","checked":"2026-11-14"},{"url":"https://docs.tembo.io/features/sandbox/overview","quote":"Each session runs in a dedicated Linux VM, and no two sessions share the same VM.","checked":"2026-11-14"},{"url":"https://docs.tembo.io/features/sandbox/overview","quote":"Sandboxes are ephemeral: spun up for the session, destroyed when it's done.","checked":"2026-11-14"},{"url":"https://docs.tembo.io/features/sandbox/overview","quote":"Pro goes up to Medium. The Max plan goes up to XL. XXL and Ultra are available on request.","checked":"2026-11-14"},{"url":"https://docs.tembo.io/features/sandbox/overview","quote":"| Ultra | 32 | 128 GB | 200 GB |","checked":"2026-11-14"},{"url":"https://www.tembo.io","quote":"Reusable snapshots, setup and pre-run scripts, adjustable compute, and on-demand VM keep-alive","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/tembo","score":{"dev-box":{"total":58,"rank":27,"of":38},"coding-agent":{"total":76,"rank":2,"of":11}},"comparablePrice":{"dev-box":60,"coding-agent":null},"benchmarks":[]},{"slug":"tenki","name":"Tenki Sandbox","company":"Tenki Cloud (Luxor Technology)","website":"https://tenki.cloud/products/sandbox","docsUrl":"https://tenki.cloud/docs/sandbox/quickstart","pricingUrl":"https://tenki.cloud/docs/pricing","categories":["code-sandbox","ci-runner"],"tagline":"Disposable full Linux VMs for agents, billed per second, with pause, memory snapshots and Docker inside.","summary":"Tenki Sandbox runs each session in a hardware-isolated Linux VM with its own kernel, root access and Docker support, driven from a CLI, an MCP server or Go, TypeScript and Python SDKs. Sessions bill per second at $0.0504 per CPU core-hour and $0.0162 per GiB-hour of memory, with the first 5 GiB of disk free. Sessions can be paused with memory intact, snapshotted (disk and memory) and forked. Tenki also sells GitHub Actions runners from the same account and credits.","isolation":"Hardware-isolated VM per session with its own kernel","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Go","JavaScript/TypeScript","Python","CLI","MCP server"],"regions":"Not stated. Runs on Tenki-owned hardware with a Ceph storage cluster.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.0504,"gibHour":0.0162,"instanceHour":null,"instance":null,"storageGibMonth":0.2,"billingIncrement":"per second","billedWhile":"CPU cores and memory while a session is active; a paused session stops compute charges. Root disk above 5 GiB bills per second; snapshots and volumes bill from the workspace storage pool.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Starter","monthlyFee":0,"includedUsageUsd":0,"limits":"5 members, 5 concurrent sessions, up to 4 cores and 8 GB RAM per VM, 100 GiB storage. One-time $100 free credit once a payment method is added; usage beyond it billed at the listed rates.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Team","monthlyFee":250,"includedUsageUsd":100,"limits":"$250 a month billed monthly or $200 a month billed yearly; $100 of credit each month. 50 members, 100 concurrent sessions, up to 16 cores and 64 GB RAM per VM, 1,000 GiB storage, priority support.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"One-time $100 in free credits the first time a payment method is added; nothing is credited before a card is on file. Adding the card also turns on auto-reload, which adds $20 whenever the balance drops below $5, and can be turned off.","notes":["Rates converted from per-second prices: $0.000014 per vCPU-second x 3600 = $0.0504 per hour; $0.0000045 per GiB-second x 3600 = $0.0162 per hour. Tenki's product page labels the CPU unit as a core.","Root disk: $0.00000003 per GiB-second ($0.000108 per GiB-hour, about $0.079 per GiB-month) after 5 GiB free. Default root disk is 5 GiB.","Snapshots, volumes and runner caches share one workspace storage pool at $0.20 per GiB-month from the first byte (the storage rate we show).","Default session is 2 vCPU / 4096 MB. Sessions have a configurable maximum duration, or run until terminated when marked sticky; no hard cap is published.","Inbound and outbound networking are on by default and can be disabled, with outbound allowlists available.","CI runners: $0.002 per vCPU-minute for x64 Linux and $0.020 per vCPU-minute for macOS on Apple Silicon. AI code reviews cost $1.00 each. All draw from the same credits.","Templates are built from a Git context; Dockerfiles are not currently accepted."]},"bestFor":["Agents that need a full VM with root, systemd and Docker","Long agent sessions that pause with memory intact and resume later","Teams that also want cheaper GitHub Actions runners on the same bill"],"watchOut":["Starter caps each VM at 4 cores / 8 GB and 5 active sessions","Snapshot and volume storage costs $0.20 per GiB-month from the first byte","Regions are not documented","Templates do not accept Dockerfiles"],"sources":[{"url":"https://tenki.cloud/docs/pricing","quote":"Sandbox vCPU   | $0.000014 per second","checked":"2026-11-13"},{"url":"https://tenki.cloud/docs/pricing","quote":"Sandbox Memory | $0.0000045 per GiB/second","checked":"2026-11-13"},{"url":"https://tenki.cloud/docs/pricing","quote":"$0.20 per GiB/month, from the first byte","checked":"2026-11-13"},{"url":"https://tenki.cloud/docs/pricing","quote":"$100 in free credits when you add a payment method","checked":"2026-11-14"},{"url":"https://tenki.cloud/docs/account/free-credits","quote":"New Tenki accounts get $100 in free credits the first time they add a payment method.","checked":"2026-11-14"},{"url":"https://tenki.cloud/docs/account/free-credits","quote":"Adding the payment method also turns on auto-reload, which adds $20 whenever your balance drops below $5.","checked":"2026-11-14"},{"url":"https://tenki.cloud/docs/account/free-credits","quote":"The Team plan ($200/month yearly, $250/month monthly) includes $100 in credits every month","checked":"2026-11-14"},{"url":"https://tenki.cloud/docs/pricing","quote":"$200/mo billed yearly, $250/mo billed monthly","checked":"2026-11-13"},{"url":"https://tenki.cloud/docs/pricing","quote":"x64 Runners    | $0.002 per vCPU/minute","checked":"2026-11-13"},{"url":"https://tenki.cloud/products/sandbox","quote":"Billed per second. Hourly rate shown for reference. First 5 GiB of disk free.","checked":"2026-11-13"},{"url":"https://tenki.cloud/products/sandbox","quote":"Each sandbox is a disposable, hardware-isolated VM.","checked":"2026-11-13"},{"url":"https://tenki.cloud/docs/pricing","quote":"4 cores, 8 GB RAM","checked":"2026-11-14"},{"url":"https://tenki.cloud/docs/pricing","quote":"16 cores, 64 GB RAM","checked":"2026-11-14"},{"url":"https://tenki.cloud/docs/sandbox/snapshots","quote":"A snapshot is a restorable VM capture of one session, including both disk and memory state.","checked":"2026-11-13"},{"url":"https://tenki.cloud/docs/sandbox/sessions","quote":"Use the network options to disable either direction, or set an","checked":"2026-11-13"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/tenki","score":{"code-sandbox":{"total":67,"rank":27,"of":63},"ci-runner":{"total":71,"rank":4,"of":10}},"comparablePrice":{"code-sandbox":29.1456,"ci-runner":23.04},"benchmarks":[]},{"slug":"tensorlake","name":"Tensorlake Sandboxes","company":"Tensorlake","website":"https://tensorlake.ai","docsUrl":"https://docs.tensorlake.ai/sandboxes/introduction","pricingUrl":"https://tensorlake.ai/pricing","categories":["code-sandbox","dev-box","gpu"],"tagline":"Firecracker microVM sandboxes billed on active CPU time (20% floor), with suspend/resume, snapshots and SSH.","summary":"Tensorlake runs CPU sandboxes in Firecracker microVMs, created from Python, TypeScript, the tl CLI or the API. Ephemeral sandboxes run until terminated or timed out; named sandboxes can be suspended and resumed with memory, processes and filesystem kept, and can be used as SSH dev environments. CPU is billed on active execution time with a floor of 20% of allocated vCPU, while RAM and root disk are billed on allocation while the sandbox runs. Plans range from a free tier of one small sandbox to prepaid credits and a $250/month Pro plan with lower rates.","isolation":"Firecracker microVM per sandbox (GPU sandboxes use gVisor)","isolationKind":"microvm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python","TypeScript","CLI","REST API"],"regions":"Region list not published; the docs state EU data residency is supported. Enterprise can run in your own AWS, GCP or Azure account.","features":{"pauseResume":true,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":16,"maxMemoryGib":64},"pricing":{"rates":{"vcpuHour":0.07,"gibHour":0.015,"instanceHour":null,"instance":null,"storageGibMonth":0.07,"billingIncrement":"per second, billed in whole compute units ($0.01 each), pooled per organization","billedWhile":"CPU: active execution time, with a minimum of 20% of allocated vCPU (full allocation when active metering is unavailable). RAM and root disk: allocated amount while running. Suspended sandboxes pay snapshot storage only.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"NVIDIA H100 (reserved capacity, separate invoice)","perHour":3.95}],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"1 sandbox at a time, 1 vCPU, 1 GB RAM, 10 GB disk, sessions up to 2 hours. Sessions are not metered; no paid usage is possible on this plan.","includedVcpuHours":0,"includedGibHours":0,"capped":true},{"name":"Usage Credits","monthlyFee":0,"includedUsageUsd":0,"limits":"Prepaid packs of $5, $10 or $20 (500-2,000 compute units); no postpaid overage, sandboxes stop when the balance hits zero unless auto-refill is on. Up to 100 concurrent sandboxes, 4 vCPU, 16 GB RAM, 100 GB disk, sessions up to 24 hours.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Pro","monthlyFee":250,"includedUsageUsd":250,"limits":"$250/month including 25,000 compute units ($250) of usage at Pro rates, then $0.01 per unit. 1,000 concurrent sandboxes, up to 16 vCPU, 64 GB RAM, 100 GB disk, no session time limit. HIPAA with a BAA.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan at $0 with no credit card: 1 sandbox at a time with 1 vCPU, 1 GB RAM and 10 GB disk, sessions up to 2 hours, unmetered.","notes":["Rates shown are the Usage Credits (pay-as-you-go) rates: $0.07 per active CPU core-hour and $0.015 per GB-hour of RAM. Pro rates are 40% lower: $0.042 per active core-hour and $0.009 per GB-hour of RAM, so the site's Pro estimate, which applies the higher rates against Pro's $250 credit, overstates Pro cost.","Root disk is billed on allocated size while the sandbox runs: $0.0002 per GB-hour on Usage Credits, $0.0001 on Pro (about $0.146 and $0.073 per GB-month if run all month). The storage rate shown is snapshot storage, $0.07 per GB-month on both plans, billed while a sandbox is suspended. Network egress is free.","Prices are quoted per GB; the usage tab in the dashboard reports GiB-hours. We treat GB as GiB.","Memory must be between 1 GB and 8 GB per vCPU. Max size and session length depend on the plan: Free 1 vCPU / 1 GB / 2 hours, Usage Credits 4 vCPU / 16 GB / 24 hours, Pro 16 vCPU / 64 GB / no limit. The max size and session length shown are Pro's.","Only named sandboxes can be suspended and resumed; ephemeral sandboxes run until terminated or timed out. A terminated sandbox can be restarted within 48 hours.","CPU is billed on active execution but never below 20% of the allocated vCPU: an idle 8-vCPU sandbox bills 1.6 vCPU. The site's estimate does not apply this floor, so it understates cost for sandboxes whose CPU is busy less than 20% of the time.","Conflict: the pricing page states the 20% CPU minimum; the billing docs page does not mention it and says only that CPU tracks active execution.","GPU sandboxes are available by request, run on gVisor instead of VMs, and cannot be live resized. The pricing page lists reserved NVIDIA H100 capacity at $3.95 per GPU-hour, billed on its own invoice separate from sandbox usage; reservations go through sales.","Running in your own cloud account (in-VPC or on-prem) is an Enterprise option with custom pricing.","When a Usage Credits balance reaches zero, new sandbox activity is blocked and running sandboxes are stopped."]},"bestFor":["Agent tool calls and harnesses that wait on model calls, since only active CPU time is billed","Named sandboxes used as SSH or IDE dev environments that suspend when idle and resume with processes intact","Evals and RL rollouts that fork a warmed sandbox into many copies"],"watchOut":["RAM and root disk are billed on allocation for the whole run, and idle CPU still bills 20% of the allocated vCPU","Pay-as-you-go sandboxes are limited to 4 vCPU, 16 GB and 24-hour sessions; larger sizes and unlimited sessions need the $250/month Pro plan","GPU sandboxes need a request; the only published GPU price is reserved H100 capacity","Region list is not published"],"sources":[{"url":"https://tensorlake.ai/pricing","quote":"An idle 8-vCPU sandbox bills 1.6 vCPU; a busy one bills what it uses.","checked":"2026-11-14"},{"url":"https://tensorlake.ai/pricing","quote":"Includes 25,000 CU ($250 of usage) per billing cycle. Additional usage is billed at $0.01 per CU at the Pro rates below.","checked":"2026-11-14"},{"url":"https://tensorlake.ai/pricing","quote":"We block new sandbox activity and stop running sandboxes instead of turning the shortfall into an unexpected bill.","checked":"2026-11-14"},{"url":"https://tensorlake.ai/pricing","quote":"The NVIDIA H100 is $3.95 per GPU-hour.","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"Active CPU per core-hour | $0.07 | $0.042 | Custom","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"RAM per GB-hour | $0.015 | $0.009 | Custom","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"Root disk per GB-hour | $0.0002 | $0.0001 | Custom","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"Snapshot storage per GB-month | $0.07 | $0.07 | Custom","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"Rates are quoted per hour for readability, but runtime usage is measured by the second.","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"Pro | $250 / month + usage | 1,000 | 16 vCPU, 64 GB RAM, 100 GB root disk | Unlimited","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"Free | $0 forever | 1 | 1 vCPU, 1 GB RAM, 10 GB root disk | 2 hours","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/platform/billing","quote":"allocate between 1 GB and 8 GB of RAM per vCPU","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/sandboxes/introduction","quote":"GPU sandboxes use gVisor while CPU-only sandboxes run in VMs.","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/sandboxes/introduction","quote":"Tensorlake is HIPAA and SOC 2 Type II compliant, supports EU data residency, and offers zero data retention.","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/sandboxes/lifecycle","quote":"Named sandbox is paused. Consumes no compute; state is preserved for resume under the same sandbox ID.","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/sandboxes/gpu","quote":"GPU sandboxes are available by request.","checked":"2026-11-14"},{"url":"https://docs.tensorlake.ai/sandboxes/networking","quote":"Restricting the sandbox's own outbound internet access","checked":"2026-11-14"},{"url":"https://tensorlake.ai","quote":"Each tool call and harness runs in its own microVM. No shared kernel, no cross-tenant state.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/tensorlake","score":{"code-sandbox":{"total":76,"rank":5,"of":63},"dev-box":{"total":78,"rank":3,"of":38},"gpu":{"total":73,"rank":1,"of":7}},"comparablePrice":{"code-sandbox":17.951999999999998,"dev-box":19.36,"gpu":3.95},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":991,"unit":"ms","better":"lower","date":"2026-11-13","rank":16,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":72954,"unit":"ms","better":"lower","date":"2026-11-13","rank":21,"of":25}]},{"slug":"thunder-compute","name":"Thunder Compute","company":"Thunder Compute","website":"https://www.thundercompute.com","docsUrl":"https://www.thundercompute.com/docs","pricingUrl":"https://www.thundercompute.com/pricing","categories":["gpu"],"tagline":"Low-cost GPU cloud with persistent GPU instances and per-second Firecracker GPU sandboxes for agents.","summary":"Thunder Compute rents GPUs two ways. Instances are persistent GPU machines you create from a console, CLI or VS Code, billed per minute, from $0.35/hour for an RTX A6000 to $3.20/hour for an H100. Sandboxes are Firecracker microVMs for agents, RL and evals, created from a Python SDK, MCP server or REST API and billed per second for each GPU, vCPU, memory and storage. Sandbox access must be enabled for your organization on request, and there are no egress fees.","isolation":"Sandboxes: Firecracker microVM per sandbox. Instances: EC2-based machines sharing a pool of network-attached GPUs","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["Python (thunder-sandbox)","CLI (tnr)","REST API","MCP server","VS Code extension"],"regions":"Not stated","features":{"pauseResume":null,"snapshots":true,"gpu":true,"customImages":true,"networkControls":true,"maxSessionHours":null,"maxVcpu":null,"maxMemoryGib":null},"pricing":{"rates":{"vcpuHour":0.051,"gibHour":0.016,"instanceHour":null,"instance":null,"storageGibMonth":0.0876,"billingIncrement":"per second (sandboxes); per minute (instances)","billedWhile":"Sandboxes: each resource per second until the sandbox stops or its TTL ends. Instances: while running; snapshots bill until deleted.","cpuActiveOnly":false,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[{"type":"H100 SXM (sandbox)","perHour":2.95},{"type":"A100 80GB (sandbox)","perHour":1.95},{"type":"H100 SXM (instance, 4 vCPU included)","perHour":3.2},{"type":"A100 80GB (instance, 8 vCPU included)","perHour":1.09},{"type":"L40 (instance, 6 vCPU included)","perHour":0.79},{"type":"RTX A6000 (instance, 6 vCPU included)","perHour":0.35}],"plans":[],"freeTierKind":"none","freeTier":"No free tier or trial credit is stated on the pricing or billing pages.","notes":["The vCPU, memory and storage rates shown are sandbox rates; storage is $0.00012 per GiB-hour, shown per month as x 730 hours. CPU-only sandboxes pay only these rates.","Sandboxes are not open to everyone yet: access must be enabled for your organization after a request. Instances are self-serve from the console.","Instance prices are per GPU-hour and include a set number of vCPUs and RAM; extra vCPUs cost $0.04 per vCPU-hour. The first 100 GB of instance disk per GPU is included, then $0.219 per GB-month while running; instance snapshots cost $0.05 per GB-month.","There are no data egress fees. Instances and sandboxes are never preempted.","Sandbox disks are wiped when the sandbox ends; instances keep their disk and support snapshots.","Sandbox outbound internet can be left open, closed, or limited to domain and CIDR allowlists, and changed on a running sandbox.","Maximum sandbox vCPU and memory are set by a resource catalog that is not published on the pages checked; instances allow up to 16 vCPUs per GPU on A100 and H100."]},"bestFor":["Low hourly GPU prices for development, notebooks and training on persistent instances","GPU-backed RL rollouts, evals and CI from code, once sandbox access is enabled","Workloads that move a lot of data out, since egress is free"],"watchOut":["Sandbox access requires a request; only instances are fully self-serve","Instances share a pool of network-attached GPUs rather than reserving one","No free tier or trial credit is published","Regions are not stated"],"sources":[{"url":"https://www.thundercompute.com/pricing","quote":"|vCPU |$0.051 /vCPU-hr $0.0000142 /vCPU-sec |","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"|Memory |$0.016 /GiB-hr $0.00000444 /GiB-sec |","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"|Storage |$0.00012 /GiB-hr $0.0000000333 /GiB-sec |","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"|H100 SXM |$2.95 /GPU-hr $0.000819 /GPU-sec |","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"|A100 80GB |$1.95 /GPU-hr $0.000542 /GPU-sec |","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"|H100 SXM |80 GB |4 included · 16 max |","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"$0.35 /hr","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"Request access and our sales team will follow up to get you set up.","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/pricing","quote":"No, Thunder Compute does not bill for data egress.","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/docs/sandboxes/quickstart","quote":"Thunder sandboxes are isolated Ubuntu microVMs powered by Firecracker, not gVisor containers.","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/docs/sandboxes/quickstart","quote":"Sandbox access must be enabled for your organization.","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/docs/faq/faq","quote":"Instead of reserving a GPU, cloud instances on Thunder share a pool of network-attached GPUs.","checked":"2026-11-13"},{"url":"https://www.thundercompute.com/products/sandboxes","quote":"Close the network for evals so the model can’t look answers up.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/thunder-compute","score":{"gpu":{"total":50,"rank":6,"of":7}},"comparablePrice":{"gpu":2.95},"benchmarks":[]},{"slug":"ubicloud","name":"Ubicloud GitHub Actions runners","company":"Ubicloud","website":"https://www.ubicloud.com","docsUrl":"https://www.ubicloud.com/docs/github-actions-integration/quickstart","pricingUrl":"https://www.ubicloud.com/docs/about/pricing","categories":["ci-runner"],"tagline":"Managed GitHub Actions runners on Ubicloud's open-source cloud, from $0.00125 per minute for 2 vCPU.","summary":"Ubicloud runs GitHub Actions jobs in a fresh, ephemeral VM per job on its own cloud built on Cloud Hypervisor and bare-metal hosts in Germany. You connect a GitHub App and change the runs-on label; x64 and arm64 runners come in 2 to 30 vCPU sizes. Standard runners cost $0.00125 per minute for 2 vCPU and premium runners, enabled by default, $0.002, both billed by the minute. Every account gets a $2.5 monthly credit, and the platform source code is available under AGPL-3.0.","isolation":"Clean, ephemeral VM per job on Cloud Hypervisor","isolationKind":"microvm","openSource":true,"selfHost":true,"license":"AGPL-3.0","repoUrl":"https://github.com/ubicloud/ubicloud","lastRelease":"2026-11-13","sdks":["GitHub Actions runs-on labels","GitHub App"],"regions":"Managed runners are placed across data centers in Germany.","features":{"pauseResume":false,"snapshots":null,"gpu":null,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":30,"maxMemoryGib":120},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.075,"instance":"Standard x64 runner: 2 vCPU, 8 GB, Linux ($0.00125 per minute)","storageGibMonth":null,"billingIncrement":"per minute","billedWhile":"While a job's runner VM is provisioned; the VM is removed when the job ends.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Pay as you go","monthlyFee":0,"includedUsageUsd":2.5,"limits":"$2.5 monthly account credit (1,250 standard 2 vCPU runner minutes), then per-minute rates. A valid credit card is required before creating resources.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"$2.5 credit every month, equal to 1,250 minutes on the 2 vCPU standard runner. A credit card is required to start.","notes":["Rate shown is the standard x64 2 vCPU / 8 GB runner: $0.00125 per minute x 60 = $0.075 per hour. Sizes scale linearly: 4 vCPU $0.0025, 8 vCPU $0.005, 16 vCPU $0.01 and 30 vCPU $0.01875 per minute.","Premium runners (AMD Ryzen 9 7950X3D) are enabled by default and cost $0.002 per minute for 2 vCPU ($0.12 per hour); jobs fall back to standard runners when premium capacity is full and are billed at the rate of the runner used. Premium can be turned off in the console.","arm64 runners cost the same as standard x64 ($0.00125 per minute for 2 vCPU / 6 GB).","Linux only (Ubuntu 22.04, 24.04 and 26.04 images built from GitHub's runner image templates); no Windows or macOS runners are listed.","Cache: 30 GB free per repository (100 GB with premium runners); extra cache storage is $0.04 per GB-month in 25 GB steps, set up through support.","Memory is stated in GB; we treat it as GiB.","The Ubicloud platform is open source under AGPL-3.0 and can be run on your own bare-metal hosts; the prices here are for the managed service."]},"bestFor":["Teams that want the lowest per-minute price for Linux GitHub Actions jobs","Workloads that need arm64 runners at the same price as x64","Teams that prefer an open-source platform they could run themselves"],"watchOut":["Runners run only in Germany","Linux only; no macOS or Windows runners","Premium runners are on by default and cost 60% more than standard","A credit card is required even to use the free monthly credit"],"sources":[{"url":"https://www.ubicloud.com/docs/about/pricing","quote":"GitHub Actions: $0.00125/min on Ubicloud vs. $0.00600/min on GitHub","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/about/pricing","quote":"Every account gets a $2.5/month credit that's equivalent to 1,250 minutes of Ubicloud runner time.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/about/pricing","quote":"All new users get the superior performance of our premium runners by default.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/about/pricing","quote":"We calculate resource consumption at per minute granularity and bill monthly.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/about/pricing","quote":"Our GitHub Managed Runner Application allocates virtual machines (VMs) across data centers in Germany to provide high availability.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/github-actions-integration/runner-types","quote":"The supported number of vCPUs are 2 (default), 4, 8, 16, and 30.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/github-actions-integration/security","quote":"For each new CI/CD job, we also provide you with a clean and ephemeral VM.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/github-actions-integration/quickstart","quote":"Ubicloud requires an active, valid credit card on file before you can create resources.","checked":"2026-11-13"},{"url":"https://www.ubicloud.com/docs/github-actions-integration/ubicloud-cache","quote":"Ubicloud Cache offers 30 GB of free storage per repository","checked":"2026-11-13"},{"url":"https://github.com/ubicloud/ubicloud","quote":"Linux VMs on Cloud Hypervisor, with public IPv4 and IPv6, private networking, and encrypted disks.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/ubicloud","score":{"ci-runner":{"total":79,"rank":1,"of":10}},"comparablePrice":{"ci-runner":7.5},"benchmarks":[]},{"slug":"upstash-box","name":"Upstash Box","company":"Upstash","website":"https://upstash.com/docs/box","docsUrl":"https://upstash.com/docs/box/overall/how-it-works","pricingUrl":"https://upstash.com/pricing/box","categories":["code-sandbox"],"tagline":"Durable agent containers billed per active CPU hour, auto-paused when idle, with built-in coding agents and browser.","summary":"Upstash Box gives each agent an isolated container with a shell, filesystem, git, snapshots, an optional managed Chromium and built-in coding agents such as Claude Code, Codex and OpenCode. Standard boxes bill only for CPU actually consumed, at $0.10 per core-hour, and freeze automatically when idle while keeping their storage; disk is $0.10 per GB-month. Keep-alive boxes stay on continuously for a flat $0.0111 to $0.0444 per hour by size. A free tier gives 10 concurrent boxes and 5 CPU hours a month.","isolation":"Isolated container per box with its own filesystem, shell and network stack","isolationKind":"container","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","Python","MCP server"],"regions":"AWS us-east-1 only; more regions planned.","features":{"pauseResume":true,"snapshots":true,"gpu":null,"customImages":false,"networkControls":true,"maxSessionHours":null,"maxVcpu":8,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":0.1,"gibHour":0,"instanceHour":null,"instance":null,"storageGibMonth":0.1,"billingIncrement":"by CPU time consumed, in core-hours","billedWhile":"Standard boxes: only for CPU actually consumed; idle boxes with no CPU use cost nothing for compute and freeze automatically. Storage bills while data exists. Keep-alive boxes bill a flat hourly rate for as long as they exist.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Free","monthlyFee":0,"includedUsageUsd":0,"limits":"Up to 10 concurrent boxes and 5 CPU hours per month. Agents use your own model provider key.","includedVcpuHours":5,"includedGibHours":0,"capped":true},{"name":"Pay as You Go","monthlyFee":0,"includedUsageUsd":0,"limits":"Up to 1,000 concurrent boxes (soft limit, can be raised). Standard boxes at $0.10 per active CPU hour plus $0.10 per GB-month of storage; keep-alive boxes at a fixed hourly rate.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Free plan: 5 CPU hours per month and up to 10 concurrent boxes.","notes":["Active CPU is measured in core-hours: 100% of 2 cores for an hour costs $0.20, and 10% of one core for an hour costs $0.01. Memory is not billed separately (we show $0 per GiB-hour).","Sizes: Small 2 vCPU / 4 GB, Medium 4 vCPU / 8 GB, Large 8 vCPU / 16 GB. The size table lists 10/30/30 GB disk, while the FAQ lists storage limits of 10/20/30 GB.","Keep-alive boxes never pause and cost $0.0111/hour ($8/month) Small, $0.0222/hour ($16/month) Medium, $0.0444/hour ($32/month) Large, billed per second, covering CPU and storage.","Storage is $0.10 per GB-month and includes snapshots.","Idle standard boxes freeze: storage is kept but running processes stop until the next request wakes the box.","The pricing page does not say what happens when the free plan's 5 CPU hours run out; we treat the free plan as having no overage billing.","Custom Docker containers are planned but not yet available. Egress can be set to deny-all or an allowlist per box.","Model usage for built-in agents is billed by your own model provider, not Upstash."]},"bestFor":["Agents that mostly wait, since standard boxes bill only for CPU actually used","Always-on agent boxes at a low flat monthly price with keep-alive","Running Claude Code, Codex or OpenCode in a box with your own model key"],"watchOut":["Container isolation rather than a VM","Only one region (AWS us-east-1)","No custom Docker images yet","Idle boxes freeze and stop their running processes"],"sources":[{"url":"https://upstash.com/pricing/box","quote":"| Small | 2 vCPU | 4 GB RAM | 10 GB | $0.1/active CPU hour | $0.0111/hour ($8/month) |","checked":"2026-11-13"},{"url":"https://upstash.com/pricing/box","quote":"For example, using 100% of 2 cores for one hour costs $0.2, while using 10% of a single core for one hour costs $0.01.","checked":"2026-11-13"},{"url":"https://upstash.com/pricing/box","quote":"$0.10 per GB/month (billed separately on Pay as You Go)","checked":"2026-11-13"},{"url":"https://upstash.com/pricing/box","quote":"The free tier includes up to 10 concurrent boxes and 5 CPU hours per month.","checked":"2026-11-13"},{"url":"https://upstash.com/pricing/box","quote":"Boxes currently run on AWS us-east-1. We are planning to add more regions soon.","checked":"2026-11-13"},{"url":"https://upstash.com/pricing/box","quote":"Yes, custom Docker container support is planned and coming soon.","checked":"2026-11-13"},{"url":"https://upstash.com/pricing/box","quote":"| Large | 8 vCPU | 16 GB RAM | 30 GB | $0.4/active CPU hour | $0.0444/hour ($32/month) |","checked":"2026-11-13"},{"url":"https://upstash.com/docs/box/overall/how-it-works","quote":"Each box is an isolated container with its own filesystem, shell, network stack, and optional coding agent.","checked":"2026-11-13"},{"url":"https://upstash.com/docs/box/overall/how-it-works","quote":"Capture box state and restore new boxes from it","checked":"2026-11-13"},{"url":"https://upstash.com/docs/box/guides/eve-setup","quote":"Egress is deny-all by default, setup is baked into a snapshot at build time, and each session gets its own box.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/upstash-box","score":{"code-sandbox":{"total":73,"rank":11,"of":63}},"comparablePrice":{"code-sandbox":10.56},"benchmarks":[]},{"slug":"vercel-sandbox","name":"Vercel Sandbox","company":"Vercel","website":"https://vercel.com/sandbox","docsUrl":"https://vercel.com/docs/vercel-sandbox","pricingUrl":"https://vercel.com/docs/vercel-sandbox/pricing","categories":["code-sandbox"],"tagline":"Firecracker microVMs on Vercel, billed for active CPU time plus provisioned memory.","summary":"Vercel Sandbox runs each session in a Firecracker microVM, created from the JavaScript/TypeScript or Python SDK or the sandbox CLI. CPU is billed only for time the code actively uses it, while memory is billed for the full provisioned amount for as long as the session runs (2 GB per vCPU). Sandboxes are persistent by default: stopping one snapshots its filesystem (billed as snapshot storage) and the next call resumes it in a new session of up to 24 hours on Pro. Hobby includes a monthly usage allowance at no cost; Pro usage draws from the plan's $20 monthly credit.","isolation":"Firecracker microVM per sandbox","isolationKind":"microvm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["JavaScript/TypeScript","Python"],"regions":"19 Vercel regions, default iad1 (Washington, D.C.); region chosen per sandbox or per project. Rates for CPU, memory, data transfer and drives vary by region.","features":{"pauseResume":true,"snapshots":true,"gpu":false,"customImages":true,"networkControls":true,"maxSessionHours":24,"maxVcpu":8,"maxMemoryGib":16},"pricing":{"rates":{"vcpuHour":0.128,"gibHour":0.0212,"instanceHour":null,"instance":null,"storageGibMonth":0.08,"billingIncrement":"Active CPU by CPU time used; provisioned memory in 1-minute minimum increments","billedWhile":"CPU: only while code actively uses it (I/O and model-call waits are free). Memory: full provisioned amount for the whole session, including idle time. Nothing while stopped, except snapshot storage.","cpuActiveOnly":true,"unitVcpu":null,"unitGib":null,"browserHour":null},"gpu":[],"plans":[{"name":"Hobby","monthlyFee":0,"includedUsageUsd":0,"limits":"Monthly allowance (not a dollar credit): 5 Active CPU hours, 420 GB-hours of memory, 5,000 sandbox creations, 20 GB data transfer; 15 GB snapshot and 15 GB drive storage (lifetime). 10 concurrent sandboxes, 45-minute sessions, max 4 vCPU / 8 GB. Sandbox creation pauses when the allowance is used up.","includedVcpuHours":5,"includedGibHours":420,"capped":true},{"name":"Pro","monthlyFee":20,"includedUsageUsd":20,"limits":"$20/month platform fee (1 deploying seat) with $20/month usage credit shared across all Vercel usage. 10,000 concurrent sandboxes, 24-hour sessions, max 8 vCPU / 16 GB, vCPU allocation up to 5,000 per minute.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"recurring","freeTier":"Hobby: free monthly allowance of 5 Active CPU hours, 420 GB-hours of memory, 5,000 creations and 20 GB data transfer; no overage billing, creation pauses until the next cycle.","notes":["Rates shown are Pro/Enterprise rates for the default iad1 region: $0.128 per Active CPU hour and $0.0212 per GB-hour of provisioned memory. Other regions have different rates.","Ambiguity: Vercel states memory in GB (GB-hour) without saying GB or GiB; we use the GB-hour rate as the GiB-hour rate.","Active CPU sums CPU time across all vCPUs, so the vCPU-hour rate applies to vCPU-hours of actual CPU use, not wall-clock vCPU allocation. Memory is billed for provisioned size (2 GB per vCPU) for the whole run, in 1-minute minimum increments.","Sandbox creations cost $0.60 per million. Data transfer (outbound plus all exposed-port traffic) is included in the Flat Rate CDN on Pro and $0.15/GB on Enterprise; downloads into the sandbox are free.","Snapshot storage is $0.08/GB-month (the storage rate we show). Persistent sandboxes (the default) snapshot automatically on stop, and snapshots keep incurring storage until they expire (30 days after last use by default) or are deleted. Drive storage is $0.05/GB-month, plus $0.0015/GB reads and $0.004/GB writes.","Pause/resume restores the filesystem only (a new VM session boots from the snapshot); running processes and memory are not preserved.","Enterprise allows up to 32 vCPU / 64 GB per sandbox; the max size we show is the self-serve Pro limit (8 vCPU / 16 GB). Enterprise pricing is custom.","Pro's $20 credit is shared with all other Vercel products used by the team."]},"bestFor":["Agent loops that mostly wait on LLM or API calls, since idle CPU is not billed.","Teams already on Vercel that want sandboxes under the same account, credit and regions.","Workloads needing egress policies with domain allowlists, credential injection or deny-all, changeable on a live sandbox."],"watchOut":["Memory is billed for the full provisioned amount for the whole session, even when idle.","Hobby sessions are capped at 45 minutes and usage stops (no overage) once the monthly allowance is used.","Max 8 vCPU / 16 GB per sandbox on Pro; larger sizes need Enterprise.","Persistent sandboxes create snapshots on every stop that are billed as storage until they expire or are deleted."],"sources":[{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Sandbox Active CPU | 5 hours/month | $0.128/hour","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Sandbox Provisioned Memory | 420 GB-hours/month | $0.0212/GB-hour","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Snapshot Storage | 15 GB (lifetime) | $0.08/GB-month","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Provisioned memory is billed in 1 minute minimum increments to account for sandbox lifecycle management.","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Max Session Duration | 45 minutes | 24 hours | 24 hours","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"All Sandbox usage on Pro plans is charged against your $20/month credit.","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Time spent waiting for I/O (such as network requests, database queries, or AI model calls) does not count toward Active CPU.","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox/pricing","quote":"Each vCPU includes 2 GB of memory.","checked":"2026-11-14"},{"url":"https://vercel.com/docs/plans/pro-plan","quote":"$20/month Pro platform fee","checked":"2026-11-14"},{"url":"https://vercel.com/docs/vercel-sandbox","quote":"Each sandbox runs in a secure Firecracker microVM with its own filesystem and network.","checked":"2026-11-14"},{"url":"https://vercel.com/docs/sandbox/concepts/persistent-sandboxes","quote":"Each automatic snapshot consumes Snapshot Storage, which is billed separately from compute.","checked":"2026-11-14"},{"url":"https://vercel.com/docs/sandbox/concepts/firewall","quote":"Network firewall allows users to restrict egress traffic from their sandbox.","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/vercel-sandbox","score":{"code-sandbox":{"total":76,"rank":6,"of":63}},"comparablePrice":{"code-sandbox":28.4416},"benchmarks":[{"source":"computesdk","metric":"Burst start time (median of 100 at once)","value":428,"unit":"ms","better":"lower","date":"2026-11-13","rank":9,"of":27},{"source":"computesdk","metric":"Repo clone, install and typecheck (one run)","value":72265,"unit":"ms","better":"lower","date":"2026-11-13","rank":20,"of":25},{"source":"starsling","metric":"Node.js web tooling (runs per second)","value":12.96,"unit":"runs/s","better":"higher","date":"2026-09-29","rank":10,"of":12},{"source":"starsling","metric":"Better Auth cold dependency install","value":20.7,"unit":"s","better":"lower","date":"2026-09-29","rank":10,"of":11}]},{"slug":"warpbuild","name":"WarpBuild","company":"WarpBuild","website":"https://www.warpbuild.com","docsUrl":"https://www.warpbuild.com/docs/ci/cloud-runners","pricingUrl":"https://www.warpbuild.com/pricing","categories":["ci-runner"],"tagline":"Per-minute GitHub Actions runners on ephemeral VMs, from $0.004 a minute for 2 vCPU x64, or in your own cloud.","summary":"WarpBuild provides GitHub Actions runners on ephemeral VMs that are created for each job and destroyed afterwards. Hosted cloud runners cover Linux x64 and arm64 (2 to 32 vCPU), Windows and macOS on Apple M4/M5 Pro, priced per minute with no base fee; a 2 vCPU x64 runner is $0.004 a minute. A bring-your-own-cloud option runs runners in your AWS, GCP or Azure account for a $0.002 per-minute fee plus your cloud bill. Add-ons include cache storage, snapshot runners and remote Docker builders.","isolation":"Ephemeral VM per job, destroyed when the workflow completes","isolationKind":"vm","openSource":false,"selfHost":true,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["GitHub Actions runs-on labels","REST API"],"regions":"Not stated for the hosted cloud; enterprise plans can pin runners to a chosen region, and the bring-your-own-cloud option runs in your own AWS, GCP or Azure region.","features":{"pauseResume":false,"snapshots":true,"gpu":false,"customImages":null,"networkControls":null,"maxSessionHours":null,"maxVcpu":32,"maxMemoryGib":128},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.24,"instance":"Linux x86-64 cloud runner: 2 vCPU, 8 GB ($0.004 per minute)","storageGibMonth":0.2,"billingIncrement":"per minute","billedWhile":"While a job runs on the runner VM; the VM and its disk are deleted when the job ends. Cache and snapshot storage are billed separately.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":8,"browserHour":null},"gpu":[],"plans":[{"name":"Cloud runners","monthlyFee":0,"includedUsageUsd":0,"limits":"Usage-based with no base, platform or seat fee. Unlimited job concurrency. Add-ons: cache storage $0.20/GB-month, cache operations $0.0001 each, snapshot restore $0.04 per job, snapshot storage $0.025 per hour per snapshot; SAML SSO and data residency $250/month each, audit logs $400/month.","includedVcpuHours":0,"includedGibHours":0,"capped":false},{"name":"Bring your own cloud","monthlyFee":0,"includedUsageUsd":0,"limits":"Runners in your own AWS, GCP or Azure account for $0.002 per minute (Linux, and Windows on AWS and Azure), plus the compute bill from your cloud provider. Cache and snapshots included.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"One-time $10 credit on signup, no credit card required; a payment method is needed once the free credit is used.","notes":["Rate shown is the Linux x86-64 2 vCPU / 8 GB runner: $0.004 per minute x 60 = $0.24 per hour. Linux sizes scale linearly up to 32 vCPU / 128 GB at $0.064 per minute.","Linux arm64 is cheaper: $0.003 per minute for 2 vCPU / 8 GB ($0.18 per hour). Windows starts at 4 vCPU for $0.016 per minute and macOS M4/M5 Pro at 6 vCPU for $0.08 per minute.","Bring-your-own-cloud runners cost $0.002 per minute in WarpBuild fees ($0.12 per hour) on top of your own cloud compute, which is not included in that figure.","Snapshot runners can save and restore a runner's state between jobs; restores cost $0.04 per job and stored snapshots $0.025 per hour each.","Remote Docker builders are billed separately, from $0.06 per minute for 16 vCPU / 32 GB.","Billing runs at the end of each month, but an account can be charged mid-cycle once a usage threshold is crossed. Promotional codes on some WarpBuild pages give larger one-time credits.","Memory is stated in GB; we treat it as GiB. GPU runners are listed as coming soon."]},"bestFor":["Teams that want cheaper GitHub Actions minutes with no base fee","Organizations that want runners inside their own AWS, GCP or Azure account","Jobs with long setup that can resume from a saved runner snapshot"],"watchOut":["Smallest Windows runner is 4 vCPU at $0.016 per minute","Cache storage, cache operations and snapshots are billed on top of minutes","Hosted cloud regions are not published","Free usage is a one-time $10 credit, not a monthly allowance"],"sources":[{"url":"https://www.warpbuild.com/pricing","quote":"2vCPU, 8GB RAM | $0.004","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/pricing","quote":"CI runners are billed on a per-minute basis","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/pricing","quote":"Cache Storage | $0.20 per GB-month","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/pricing","quote":"Linux Runners | $0.002/min","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/pricing","quote":"Snapshot Storage | $0.025/hour per snapshot","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/pricing","quote":"You will be need to add a payment method when you hit the free usage or credit limits.","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/answers/how-much-do-github-actions-runners-cost","quote":"Signup includes $10 free credits, so the first few thousand Linux minutes cost nothing.","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/answers/how-much-do-github-actions-runners-cost","quote":"There is no base subscription fee, no platform fee, and no seat fee.","checked":"2026-11-13"},{"url":"https://www.warpbuild.com/docs/ci/cloud-runners","quote":"All WarpBuild runners are run on ephemeral VMs for maximum isolation and security.","checked":"2026-11-13"}],"lastChecked":"2026-11-13","added":"2026-11-13","page":"https://aisandboxhunt.com/warpbuild","score":{"ci-runner":{"total":66,"rank":7,"of":10}},"comparablePrice":{"ci-runner":24},"benchmarks":[]},{"slug":"workpaper","name":"Workpaper","company":"Workpaper (individual developer, India)","website":"https://workpaper.co","docsUrl":"https://docs.workpaper.co/","pricingUrl":"https://docs.workpaper.co/pricing","categories":["dev-box","code-sandbox"],"tagline":"Small Linux VMs in India with Claude Code and Codex preinstalled, from $0.010 an hour on a $15 plan.","summary":"Workpaper starts a Linux virtual machine with its own kernel from a CLI or JavaScript and Python packages, with Claude Code, Codex and a headless browser installed. There are two sizes: small (2 vCPU, 4 GB, 10 GB disk) and large (4 vCPU, 8 GB, 20 GB disk with Docker) at twice the price. Stopping saves disk and memory, including running programs, and a computer can be forked into an independent copy. Computers pause after an hour idle, and the whole service runs on its own small pool of hardware in Delhi.","isolation":"Virtual machine with its own Linux kernel per computer (hypervisor not named)","isolationKind":"vm","openSource":false,"selfHost":false,"license":null,"repoUrl":null,"lastRelease":null,"sdks":["CLI","JavaScript/TypeScript","Python"],"regions":"India only (Delhi). The docs expect about 150 ms per keystroke from Europe and 250 ms from the US.","features":{"pauseResume":true,"snapshots":null,"gpu":false,"customImages":false,"networkControls":null,"maxSessionHours":null,"maxVcpu":4,"maxMemoryGib":8},"pricing":{"rates":{"vcpuHour":null,"gibHour":null,"instanceHour":0.01,"instance":"Small: 2 vCPU, 4 GB RAM, 10 GB disk (Pro rate). Large (4 vCPU, 8 GB, 20 GB disk, Docker) is $0.020/hour, two small units.","storageGibMonth":null,"billingIncrement":"per second","billedWhile":"Only while a computer is running. Stopped computers cost nothing, including their saved disk and memory.","cpuActiveOnly":false,"unitVcpu":2,"unitGib":4,"browserHour":null},"gpu":[],"plans":[{"name":"Pro","monthlyFee":15,"includedUsageUsd":15,"limits":"$15/month, all returned as usage credit: 1,500 small-computer hours or 750 large-computer hours a month. Unused monthly credit expires at renewal. Extra usage packs are $15 for 1,500 hours and never expire. Up to 4 running computers per account. New Pro sign-ups go to a waitlist when the service is at capacity.","includedVcpuHours":0,"includedGibHours":0,"capped":false}],"freeTierKind":"credit","freeTier":"Free trial with no account and no card: 60 minutes on one small computer, with internet limited to code and package sites and AI model APIs; one trial per device every 30 days. New accounts also get one free hour of credit with full internet.","notes":["The rate shown is the Pro rate of $0.010 per small-computer hour. Without a subscription, pay as you go is $5 once for 150 small-computer hours (75 large), about $0.033 an hour; that credit never expires. The site's estimate uses the Pro rate, so it understates pay-as-you-go cost by about 3.3 times.","Prices in India are ₹1 an hour (small) and ₹2 an hour (large).","Sizes are quoted in GB without saying GB or GiB; we treat them as GiB.","Conflict: the Limits page says there is no time limit per start and computers pause after an hour idle, while the How it works page still says computers stop 'after an hour of running'. We show no session limit. Either way, stopping keeps disk and memory and starting again resumes.","Starts depend on free host capacity. When it is full the CLI waits in line without charging; paying users go before free trials, and a running free computer may be stopped to make room.","Optional 'Instant' mode bills CPU usage at $0.006 per CPU-hour on Pro or $0.020 on pay as you go, after 50 or 10 included CPU-hours a month, plus the VM rate while a full VM runs.","No incoming connections: ports cannot be exposed and no website can be hosted on a computer.","No backups: files can be lost if hardware fails. No uptime guarantee.","Software beyond the base image (Debian 13, Python, Node.js, Claude Code, Codex, a headless browser) must be installed per computer; there are no custom images. Forking copies a set-up computer, including memory and running programs."]},"bestFor":["Cheap short agent sessions on a small VM with coding agents already installed","Users in or near India, where latency is lowest","Trying several fixes in parallel by forking a running computer, memory included"],"watchOut":["Runs only in India on a small, self-owned hardware pool; you may wait in line for a free computer","No backups and no uptime guarantee; the docs advise against data you cannot afford to lose","No inbound ports, no GPUs and no custom images","Operated by an individual developer; new Pro sign-ups can be waitlisted when at capacity"],"sources":[{"url":"https://docs.workpaper.co/pricing","quote":"Small (default): 2 vCPU · 4 GB · 10 GB disk | ₹1 an hour | $0.010 an hour","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/pricing","quote":"Billed by the second, only while a computer is running. Stopped computers cost nothing.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/pricing","quote":"is $15 a month, and all of it comes back as usage credit: 1,500 hours of the small size, or 750 of the large.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/pricing","quote":"$5 once for 150 hours of the small size, or 75 of the large (about ₹3 or $0.033 an hour for small)","checked":"2026-11-14"},{"url":"https://workpaper.co","quote":"Large · 4 vCPU, 8 GB RAM, 20 GB disk and Docker · $0.020 /hour.","checked":"2026-11-14"},{"url":"https://workpaper.co","quote":"Up to 4 running VMs per account.","checked":"2026-11-14"},{"url":"https://workpaper.co","quote":"$0.006/CPU-hour Pro · $0.020/CPU-hour PAYG","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/limits","quote":"No limit. A computer pauses after an hour idle and keeps everything; start it again to continue.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/limits","quote":"4 running computers per account, shared across small and large sizes; you can have more that are stopped.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/limits","quote":"When we're at capacity, new subscribers join a waitlist in the console","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/how-it-works","quote":"When a computer stops, we save its disk and its memory. That includes programs that were running.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/how-it-works","quote":"On our own servers in Delhi, India.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/how-it-works","quote":"Computers stop when you run workpaper stop , after an hour of running","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/free-trial","quote":"60 minutes on one computer, no account needed.","checked":"2026-11-14"},{"url":"https://docs.workpaper.co/free-trial","quote":"Every new account also comes with a free hour of credit, with full internet.","checked":"2026-11-14"},{"url":"https://workpaper.co/terms","quote":"an individual based in India who runs Workpaper","checked":"2026-11-14"}],"lastChecked":"2026-11-14","added":"2026-11-13","page":"https://aisandboxhunt.com/workpaper","score":{"dev-box":{"total":69,"rank":12,"of":38},"code-sandbox":{"total":67,"rank":28,"of":63}},"comparablePrice":{"dev-box":15,"code-sandbox":15},"benchmarks":[]}]}