NVIDIA NemoClaw
https://github.com/NVIDIA/NemoClaw
NVIDIA NemoClaw is an open-source (Apache-2.0) reference stack for running supported AI agents more safely inside NVIDIA OpenShell sandboxes. It supports OpenClaw (the default), Hermes, and LangChain Deep Agents Code, and provides guided onboarding, managed inference, network policy, managed integrations, snapshots, and lifecycle operations through the NemoClaw CLI. Developers run it on their own hosts; NVIDIA lists Linux and DGX Spark as the primary tested platforms.
| Pricing model | Open-source (Apache-2.0) self-hosted software; no product-level billing. |
|---|---|
| Pricing The headline compute price as the provider states it. Sorting ranks prices stated per vCPU over time, converted to hourly. Other pricing models sort after, unranked. | No published headline rate (open-source/self-hosted software) |
| Free while idle Whether a stopped, paused, or sleeping sandbox costs nothing. | – |
| Elastic Resources flex with what the sandbox actually uses, and the bill follows: yes when the provider bills on active or observed use, no when capacity is reserved or allocated and billed while it runs. | – |
| Memory snapshots Whether RAM state survives a pause and resume, not just disk. | – |
| Wake on request Whether a stopped sandbox wakes automatically on inbound traffic. | – |
| Start / resume Typical time from create or resume to running. Sorting ranks stated times, converted to seconds; claims with no figure sort after, unranked. | – |
| Max runtime The longest a sandbox may run. | – |
| Isolation The isolation technology between sandboxes. | – |
| GPUs Whether GPU instances are available for sandboxes. | – |
| Docker Whether Docker containers can run inside the sandbox. | – |
| SDKs | – |
✓ yes · ✕ no · – no cited public fact.
NemoClaw is an alpha project; maintainers review issues, discussions, and pull requests on a best-effort basis without guaranteed response timelines.
Sources
Last verified 2026-09-05. Corrections land through the update log.