Proper installation guide | Friends of the Crustacean 🦞🤝 | Page 1

opal galeBOT Feb 17, 2026, 8:08 PM

#

To help others find answers, you can mark your question as solved via Right click solution message -> Apps -> ✅ Mark Solution

nova tide Feb 17, 2026, 8:11 PM

#

Proper installation guide

#

@smoky hemlock help!

smoky hemlockBOT Feb 17, 2026, 8:31 PM

#

Best practice for your Windows + 3090 Ti box: run the OpenClaw Gateway inside WSL2 (Ubuntu), and run LM Studio on Windows (so it uses your RTX GPU normally). Then OpenClaw (in WSL) talks to LM Studio over HTTP.

Docs I’m basing this on:

Windows (WSL2): https://docs.openclaw.ai/platforms/windows
Getting Started: https://docs.openclaw.ai/start/getting-started
Local models (LM Studio): https://docs.openclaw.ai/gateway/local-models
FAQ (common errors): https://docs.openclaw.ai/help/faq

What you’ll end up with (mental model)

Windows: runs LM Studio + your local model + exposes http://<something>:1234/v1
WSL Ubuntu: runs OpenClaw Gateway (service) + Control UI on http://127.0.0.1:18789

#Proper installation guide

What you’ll end up with (mental model)

Step 1 — Install WSL2 + Ubuntu