Home / Compare / Isorun vs Tensorlake Sandboxes
Isorun vs Tensorlake Sandboxes
Verdict (2026-11-15): Isorun scores higher: 78/100 vs 76/100 as AI agent sandboxes. Choose Isorun for short code-execution runs from any Docker image with per-second billing and no minimum. Choose Tensorlake Sandboxes for agent tool calls and harnesses that wait on model calls, since only active CPU time is billed.
Isorun
Highest scoreKVM microVMs from any OCI image, billed per second for vCPU and RAM, with hibernate, fork and persistent disks.
$19.36 a month
Tensorlake Sandboxes
Firecracker microVM sandboxes billed on active CPU time (20% floor), with suspend/resume, snapshots and SSH.
$17.95 a month
Score by part
Cost at four workloads
| Workload | Isorun | Tensorlake Sandboxes |
|---|---|---|
| Short tasks: 1 vCPU, 1 GiB, 50 h/month | $2.00 | $1.80 |
| Dev agent: 2 vCPU, 4 GiB, 176 h/month | $19.36 | $17.95 |
| Fleet: 2 vCPU, 4 GiB, 1,760 h/month | $193.60 | $179.52 |
| Always on: 4 vCPU, 8 GiB, 730 h/month | $160.60 | $189.80 |
Per month, cheapest plan for each, without storage, network or GPU. Lowest cost in bold.
Choose Isorun if
- 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.
Choose Tensorlake Sandboxes if
- 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
Side by side
| Isorun | Tensorlake Sandboxes | |
|---|---|---|
| Category | agent sandbox, dev box | agent sandbox, dev box, GPU sandbox |
| Our score | 78/100 | 76/100 |
| Isolation | KVM microVM per sandbox with its own guest kernel | Firecracker microVM per sandbox (GPU sandboxes use gVisor) |
| vCPU per hour | $0.025 | $0.07 (active CPU only) |
| Memory per GiB-hour | $0.015 | $0.015 |
| Storage per GiB-month | $0.02 | $0.07 |
| Billing | per second, no minimums | per second, billed in whole compute units ($0.01 each), pooled per organization |
| Free tier | $50 of free credit for new accounts, no card required (about 1,250 hours of the default 1 vCPU / 1 GiB sandbox). | 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. |
| Pause and resume | Yes | Yes |
| Snapshots | Yes | Yes |
| GPU | Not stated | Yes |
| Custom images | Yes | Yes |
| Network controls | Yes | Yes |
| Max vCPU | 8 vCPU | 16 vCPU |
| Max memory | 16 GiB | 64 GiB |
| Open source | No | No |
| Self-host | No | Yes |
| SDKs and interfaces | TypeScript, MCP server, REST API | Python, TypeScript, CLI, REST API |
| Regions | US or EU; each organization is scoped to one region chosen at sign-up. | Region list not published; the docs state EU data residency is supported. Enterprise can run in your own AWS, GCP or Azure account. |
| Facts checked | 2026-11-14 | 2026-11-14 |
Isorun alternatives · Tensorlake Sandboxes alternatives
Sources
- Isorun: https://docs.isorun.ai/getting-started/pricing, checked 2026-11-14:
vCPU |$0.025 / hour
- Isorun: https://docs.isorun.ai/getting-started/pricing, checked 2026-11-14:
RAM |$0.015 / GiB / hour
- Isorun: https://docs.isorun.ai/getting-started/pricing, checked 2026-11-14:
8 vCPU + 16 GiB |$0.440 / hour
- Isorun: https://docs.isorun.ai/getting-started/pricing, checked 2026-11-14:
Every new account gets $50 in free credit, about 1,250 hours of a default sandbox. No card required to start.
- Isorun: https://docs.isorun.ai/getting-started/pricing, checked 2026-11-14:
Hibernated sandboxes: not billed for active runtime; you pay nothing while the sandbox is paused.
- Isorun: https://isorun.ai, checked 2026-11-14:
Public beta, $50 free credit.
- Isorun: https://isorun.ai, checked 2026-11-14:
Billed per second at $0.04 /hour for 1 vCPU + 1 GB RAM.
- Isorun: https://isorun.ai, checked 2026-11-14:
Simple pricing: pay for compute, egress traffic and platform fees are included.
- Isorun: https://docs.isorun.ai/sandboxes/persistent-storage, checked 2026-11-14:
$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.
- Isorun: https://docs.isorun.ai/sandboxes/lifecycle, checked 2026-11-14:
Every sandbox has an idle timeout (default 300 seconds, configurable per-create via `timeoutSec` ). When the timer fires, the sandbox is destroyed.
- Isorun: https://docs.isorun.ai/sandboxes/lifecycle, checked 2026-11-14:
It boots a microVM with its own kernel.
- Isorun: https://docs.isorun.ai/getting-started/authentication, checked 2026-11-14:
the organization is scoped to a region (US or EU)
- Tensorlake Sandboxes: https://tensorlake.ai/pricing, checked 2026-11-14:
An idle 8-vCPU sandbox bills 1.6 vCPU; a busy one bills what it uses.
- Tensorlake Sandboxes: https://tensorlake.ai/pricing, checked 2026-11-14:
Includes 25,000 CU ($250 of usage) per billing cycle. Additional usage is billed at $0.01 per CU at the Pro rates below.
- Tensorlake Sandboxes: https://tensorlake.ai/pricing, checked 2026-11-14:
We block new sandbox activity and stop running sandboxes instead of turning the shortfall into an unexpected bill.
- Tensorlake Sandboxes: https://tensorlake.ai/pricing, checked 2026-11-14:
The NVIDIA H100 is $3.95 per GPU-hour.
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
Active CPU per core-hour | $0.07 | $0.042 | Custom
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
RAM per GB-hour | $0.015 | $0.009 | Custom
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
Root disk per GB-hour | $0.0002 | $0.0001 | Custom
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
Snapshot storage per GB-month | $0.07 | $0.07 | Custom
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
Rates are quoted per hour for readability, but runtime usage is measured by the second.
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
Pro | $250 / month + usage | 1,000 | 16 vCPU, 64 GB RAM, 100 GB root disk | Unlimited
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
Free | $0 forever | 1 | 1 vCPU, 1 GB RAM, 10 GB root disk | 2 hours
- Tensorlake Sandboxes: https://docs.tensorlake.ai/platform/billing, checked 2026-11-14:
allocate between 1 GB and 8 GB of RAM per vCPU
- Tensorlake Sandboxes: https://docs.tensorlake.ai/sandboxes/introduction, checked 2026-11-14:
GPU sandboxes use gVisor while CPU-only sandboxes run in VMs.
- Tensorlake Sandboxes: https://docs.tensorlake.ai/sandboxes/introduction, checked 2026-11-14:
Tensorlake is HIPAA and SOC 2 Type II compliant, supports EU data residency, and offers zero data retention.
- Tensorlake Sandboxes: https://docs.tensorlake.ai/sandboxes/lifecycle, checked 2026-11-14:
Named sandbox is paused. Consumes no compute; state is preserved for resume under the same sandbox ID.
- Tensorlake Sandboxes: https://docs.tensorlake.ai/sandboxes/gpu, checked 2026-11-14:
GPU sandboxes are available by request.
- Tensorlake Sandboxes: https://docs.tensorlake.ai/sandboxes/networking, checked 2026-11-14:
Restricting the sandbox's own outbound internet access
- Tensorlake Sandboxes: https://tensorlake.ai, checked 2026-11-14:
Each tool call and harness runs in its own microVM. No shared kernel, no cross-tenant state.
Questions
Is Isorun or Tensorlake Sandboxes better?
Isorun scores higher on our published method: 78/100 vs 76/100 as AI agent sandboxes. Choose Isorun for short code-execution runs from any Docker image with per-second billing and no minimum. Choose Tensorlake Sandboxes for agent tool calls and harnesses that wait on model calls, since only active CPU time is billed. Checked 2026-11-14.
Is Isorun or Tensorlake Sandboxes cheaper?
Tensorlake Sandboxes is cheaper: Isorun $19.36, Tensorlake Sandboxes $17.95 a month for 2 vCPU, 4 GiB, 176 hours a month (8 h × 22 days), CPU busy 30%.
How do Isorun and Tensorlake Sandboxes isolate code?
Isorun: KVM microVM per sandbox with its own guest kernel. Tensorlake Sandboxes: Firecracker microVM per sandbox (GPU sandboxes use gVisor).