Run OpenClaw with a Local LLM
30 MIN
Install a local-first AI agent and connect it to a private OpenAI-compatible model endpoint
| Symptom | Cause | Fix |
|---|---|---|
| OpenClaw dashboard URL not loading | Gateway not running or wrong host/port | Restart the OpenClaw gateway so it reloads ~/.openclaw/openclaw.json. Verify: Check that the gateway process is running with pgrep -f openclaw or ps aux | grep openclaw. Find URL/token: Check the original installer output (scroll up in your terminal) or look in gateway logs (typically ~/.openclaw/logs/) for the dashboard URL and access token |
| "Connection refused" to model (for example localhost:8000) | Local inference server not running, still loading, or wrong port | Confirm the backend is up (curl http://localhost:8000/v1/models for vLLM, or your LM Studio / Ollama status checks) and that baseUrl in openclaw.json matches the server |
| OpenClaw says no model available | Provider not configured or model handle mismatch | Add the provider to ~/.openclaw/openclaw.json and ensure id/name exactly match the served handle |
| Out-of-memory or very slow inference | Model too large for available GPU memory or other GPU workloads | Use a smaller model from Agent-ready Models, lower context length / GPU memory utilization for vLLM, free GPU memory (close other apps), or check usage with nvidia-smi |
| WSL cannot reach model server on Windows host | localhost inside WSL is not the Windows host | Replace 127.0.0.1 / localhost in OpenClaw baseUrl with the Windows host IP as seen from WSL |
| Install script fails or dependencies missing | Missing system packages | Install curl and any required build tools; see OpenClaw documentation for current requirements |
| Config changes not applied | Gateway not reloaded | Restart the OpenClaw gateway so it reloads ~/.openclaw/openclaw.json |