NemoClaw is the runtime environment for autonomous agents—the "Matrix" where they live, work, and verify.

While coding tools like Claude help agents write logic, NemoClaw provides the infrastructure to run it, offering a programmable factory where agents can spin up physics simulations to master tasks, generate synthetic data to fix edge cases, and safely iterate through thousands of failures in isolated sandboxes.

It transforms the data center from a static deployment target into a continuous verification engine, allowing agents to autonomously build and operate complex systems—from physical robotics to self-healing infrastructure—without needing a human to manage the infrastructure.

• Docker — Docker Desktop (or a Docker daemon) must be running.
• Python 3.12+
• uv 0.9+

uv pip install nemoclaw \
  --upgrade \
  --pre \
  --index-url https://urm.nvidia.com/artifactory/api/pypi/nv-shared-pypi/simple

The nemoclaw binary is installed into your Python environment. Use uv run nemoclaw to invoke it, or activate your venv first with source .venv/bin/activate.

To install a Openclaw cluster and start a sandbox

nemoclaw sandbox create -- claude  # or opencode or codex

To run a sandbox on a remote machine, pass --remote [remote-ssh-host].

For more information see nemoclaw sandbox create --help.

The sandbox container includes the following tools by default:

For additional sandbox images see the NVIDIA/NemoClaw-Community images. You can also build your own custom images.

Note: nemoclaw sandbox create automatically deploys a cluster if one isn't already running.

To deploy a cluster explicitly:

nemoclaw cluster admin deploy

For remote deployment:

nemoclaw cluster admin deploy --remote user@host

To upgrade, redeploy your cluster to pick up the latest server and sandbox images:

nemoclaw cluster admin deploy

This will prompt you to recreate the cluster. Select "yes" to recreate the cluster.

See CONTRIBUTING.md for building from source and contributing to NemoClaw.

See architecture/ for detailed architecture docs and design decisions.