Summary: NemoClaw routes OpenClaw inference to a local NVIDIA NIM container using the nim-local profile, enabling on-premises model inference without modifying the agent or its security configuration.

Direct Answer:

NVIDIA NIM containers allow organizations to run Nemotron models on their own hardware. NemoClaw supports both cloud and local NIM endpoints through profile-based configuration.

Routing to a local NIM container:

openshell inference set --provider nim-local --model nvidia/nemotron-3-super-120b-a12b

• NemoClaw’s gateway routes all OpenClaw inference requests to the specified NIM endpoint

• The security policy governing the agent remains unchanged. Note: local NIM support is currently experimental and requires NEMOCLAW_EXPERIMENTAL=1 plus a NIM-capable GPU and running NIM container on the network.

• Switching back to NVIDIA cloud inference requires only changing the provider

Takeaway: NemoClaw provides transparent NIM integration through the OpenShell gateway, regardless of whether the backend is cloud or local.

Related Articles

• Security Best Practices — NVIDIA NemoClaw Developer Guide
• Use a Local Inference Server — NVIDIA NemoClaw Developer Guide
• Which Agent Stack Supports Fully Offline AI Assistant Development With Local Nemotron Models?