Skip to main content

Complete Guide

OpenClaw

The open-source personal AI agent with 200k+ GitHub stars. Runs on your own machine, lives in WhatsApp and Telegram, remembers everything, and takes real actions — not just suggestions. Every guide in one place.

200k+

GitHub Stars

50+

Integrations

34

Guides on this site

Any

Supported LLMs

Getting Started

What OpenClaw is and how to get it running.

Connect to Messaging Apps

Get OpenClaw running in your messaging apps of choice.

Choose Your AI Model

Connect OpenClaw to any LLM — cloud or local.

Platform Guides

Install and run OpenClaw on your specific hardware.

Comparisons

How OpenClaw compares to other AI tools.

Prompting Strategies

Get better results from OpenClaw with the right prompts and setup.

Advanced & Power User

Automation, orchestration, and power-user features.

Frequently Asked Questions

What is OpenClaw?

OpenClaw is an open-source personal AI agent with 200k+ GitHub stars. It runs on your own machine or VPS, connects to WhatsApp, Telegram, Slack, and Discord, maintains persistent local memory, and takes real actions — sending emails, running scripts, and integrating with 50+ tools.

Is OpenClaw free?

OpenClaw is free and open-source. You pay for the LLM API you connect it to (typically $5–30/month for personal use). Running local models via Ollama or LM Studio reduces API costs to zero.

Which AI model works best with OpenClaw?

Claude Sonnet is the best choice for complex tasks and reliable instruction-following. GPT-4o is strong for general use. Gemini Flash is the most cost-efficient option. For zero API costs, Llama 3.1 8B via Ollama handles everyday tasks well.

How long does OpenClaw take to set up?

On Linux or macOS, expect 1–2 hours for a full setup including WhatsApp or Telegram integration. On Windows via WSL2, add 30 minutes. On a Raspberry Pi, the base setup takes about an hour.

New to OpenClaw?

Start with the getting started guide — covers installation, your first LLM connection, and WhatsApp setup in one walkthrough.

Read the Getting Started Guide