Isorun
$19.36 a month*
Home / AI agent sandboxes / NVIDIA OpenShell
Open-source policy-enforced agent runtime from NVIDIA: sandboxes on Docker, Podman, Kubernetes or microVMs.
Verdict (2026-11-13): NVIDIA OpenShell scores 54/100, #55 of 63 AI agent sandboxes. Strong on free tier, docs and SDKs. Weak on limits. Best for: Running agents with per-sandbox file, syscall and network policies that need human review to widen.
Skip it if: Default drivers use containers; VM isolation needs the opt-in microVM driver.
Prices checked 2026-11-15.
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.
54/100 in AI agent sandboxes, rank 55 of 63. Parts with no data are left out and the rest is rescaled.
Full NVIDIA OpenShell pricing: plans and costs at four workloads →
$19.36 a month*
$11.73 a month*
$10.14 a month*
$19.00 a month*
$17.95 a month*
$28.44 a month*
$20.00 a month*
$20.00 a month*
NVIDIA OpenShell is not in the public benchmarks we track yet. See all benchmarks.
| NVIDIA OpenShell | |
|---|---|
| Category | agent sandbox |
| Our score | 54/100 |
| 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 |
| Billing | Not applicable: free software run on your own infrastructure |
| Free tier | Free open-source software under the Apache-2.0 license. |
| Pause and resume | Yes |
| GPU | Yes |
| Custom images | Yes |
| Network controls | Yes |
| Open source | Yes (Apache-2.0) |
| Self-host | Yes |
| SDKs and interfaces | 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. |
| Facts checked | 2026-11-13 |
No reviews yet. Have you used NVIDIA OpenShell? Your review helps other developers choose.
NVIDIA OpenShell scores 54/100, #55 of 63 AI agent sandboxes. Strong on free tier, docs and SDKs. Weak on limits. Best for: Running agents with per-sandbox file, syscall and network policies that need human review to widen. Facts checked 2026-11-13.
Free open-source software under the Apache-2.0 license.
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.
In AI agent sandboxes, the top-rated are Isorun (78/100), Northflank (78/100), Koyeb Sandboxes (76/100). NVIDIA OpenShell is #55 of 63.
OpenShell is the safe, private runtime for fleets of autonomous AI agents.
Agents never see real credentials; OpenShell adds them only to requests bound for approved endpoints.
This project is licensed under the Apache License 2.0.
When compute_driver is unset, the gateway auto-detects Kubernetes, then Podman, then Docker. The VM driver is never auto-detected.
The MicroVM driver runs each sandbox in its own lightweight VM. It requires host virtualization: Apple Hypervisor on macOS or KVM on Linux.
To request GPU resources, add --gpu
Stop compute when you want to retain a sandbox and its persistent workspace without keeping its container, pod, or VM running
v0.1.8