https://github.com/NVIDIA/NemoClaw - Run OpenClaw more securely inside NVIDIA OpenShell with managed inference

Gemma 4 x OpenClaw on NVIDIA Spark - Step-by-step setup guide for running Gemma 4 locally with Openclaw on NVIDIA hardware. Covers building llama.cpp with CUDA support, downloading the model via Hugging Face, configuring the local server, and connecting it to Openclaw — plus web search plugin options (Ollama or Brave) and fun demo prompts to get started quickly.