Getting Started
Install, configure, and reach your assistant in about 15 minutes
Install, configure, and reach your assistant in about 15 minutes
OpenClaw is a self-hosted personal AI assistant that runs on your own hardware. New here? Read What is OpenClaw? first, then use this guide for install, channels, and first steps (about 15 minutes).
The OpenClaw project publishes the always-current install path here: Getting started — official docs. Use the checklist below together with that page if you want both a short map and the source of truth.
Quick Start checklist
ollama launch openclaw)openclaw onboard — set LLM provider and add a channelFirst 24 hours
"tools": { "profile": "full" } in config — see FAQ: Agent doesn't run tools and Troubleshooting.openclaw backup create before changing a lot of config — see FAQ: How to back up and Workspace Backup.New to the project? Read What is OpenClaw? for a plain-language overview. Deciding between self-hosting and a cloud chatbot? See Why self-host OpenClaw? (privacy, cost, vs ChatGPT).
Follow these steps in order:
Use the one-liner, npm, or the Setup Wizard in your browser. Mac/Linux with Ollama? Fastest path: ollama launch openclaw.
Run openclaw onboard to set your LLM provider (Anthropic, OpenAI, Ollama) and add a channel (WhatsApp, Telegram, Discord).
Send a message through your connected channel or open the Control UI. You're ready.
Monthly spend ranges: How much does OpenClaw cost? Reference architecture (hardware, tiered models): Example setups & model routing.