Skip to content
Skip the setup

Want us to handle everything? Get fully managed hosting, free to start.

Back to Deploy Options

Windows PC

WSL2 or Docker Desktop

Run OpenClaw on your Windows PC using WSL2 (recommended) or Docker Desktop. Use your existing hardware for a private AI assistant.

Choose Your Method

WSL2 + Docker

Recommended

Lighter weight, better performance. Uses Windows Subsystem for Linux with native Docker.

Docker Desktop

Easier setup with GUI. Heavier resource usage but simpler management.

Pros and Cons

Advantages

  • Use existing hardware - No need to buy new equipment
  • Complete data privacy - Everything stays on your PC
  • GPU acceleration - NVIDIA GPUs can accelerate local AI models
  • Flexible messaging setup - Web chat and your preferred messaging flow can be configured on self-hosted installs
  • Familiar environment - Continue using Windows while running OpenClaw
  • No monthly costs - Free to run on your own hardware

Disadvantages

  • Not always-on by default - PC needs to be running for OpenClaw to work
  • Higher power consumption - Desktop PCs use more power than dedicated servers
  • Resource sharing - OpenClaw competes with other apps for RAM/CPU
  • WSL2 overhead - Running Linux in a VM adds some complexity
  • Windows updates - Automatic restarts can interrupt OpenClaw

Things to Consider

WSL2 vs Docker Desktop

WSL2 + Docker (Recommended): Lighter weight, better performance, integrates well with Windows. Uses ~500MB RAM overhead.
Docker Desktop: Easier GUI management, but uses more resources (~2GB RAM) and requires a paid license for commercial use in larger organizations.

Always-On Considerations

If you want 24/7 availability, configure Windows to: disable sleep mode, enable auto-login after restart, and use our startup script. Consider power costs - a gaming PC uses 100-300W, which adds up over time.

Windows Update Management

Windows updates can restart your PC unexpectedly. Configure active hours in Windows Update settings, or use Windows 11 Pro's group policy to control update behavior. The startup script ensures OpenClaw restarts automatically after reboots.

Memory Requirements

8GB total RAM is the minimum - Windows uses 3-4GB, WSL2 uses 1-2GB, leaving 2-3GB for OpenClaw. 16GB is recommended for comfortable operation alongside other apps. You can limit WSL2's memory usage in .wslconfig if needed.

Virtualization

WSL2 requires hardware virtualization (VT-x/AMD-V). Most modern PCs have this, but it may be disabled in BIOS. If you see "virtualization not enabled" errors, restart into BIOS and enable it.

Requirements

Hardware

  • • Windows 10/11 (64-bit)
  • • 8GB RAM minimum (16GB recommended)
  • • 20GB free storage
  • • Virtualization enabled in BIOS

Software

  • • Windows 10 version 2004+ or Windows 11
  • • WSL2 (we'll install this)
  • • Docker (we'll install this)

Method 1: WSL2 + Docker (Recommended)

Step 1: Install WSL2

Open PowerShell as Administrator and run:

wsl --install

Restart your computer when prompted. This installs Ubuntu by default.

Step 2: Set up Ubuntu

After restart, open Ubuntu from Start menu. Create a username and password when prompted.

Step 3: Install Docker in WSL2

In the Ubuntu terminal:

# Update packages
sudo apt update && sudo apt upgrade -y

# Install Docker
curl -fsSL https://get.docker.com | sh

# Add your user to docker group
sudo usermod -aG docker $USER

# Start Docker service
sudo service docker start

# Log out and back in (or run: newgrp docker)

Step 4: Create OpenClaw directory

mkdir -p ~/openclaw && cd ~/openclaw

Step 5: Create docker-compose.yml

cat > docker-compose.yml << 'EOF'
version: '3.8'
services:
  openclaw:
    image: openclaw/openclaw:latest
    container_name: openclaw
    restart: unless-stopped
    ports:
      - "8080:8080"
    environment:
      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY}
      - GATEWAY_TOKEN=${GATEWAY_TOKEN}
    volumes:
      - ./data:/app/data
EOF

Step 6: Create .env file

cat > .env << 'EOF'
OPENROUTER_API_KEY=sk-or-your-key-here
GATEWAY_TOKEN=$(openssl rand -hex 32)
EOF

Replace sk-or-your-key-here with your actual BYOK provider API key.

Step 7: Start OpenClaw

docker compose up -d

Access OpenClaw at http://localhost:8080 from Windows.

Method 2: Docker Desktop

Step 1: Download Docker Desktop

Download and install Docker Desktop for Windows:

Download Docker Desktop →

Step 2: Enable WSL2 backend

During installation, select "Use WSL 2 instead of Hyper-V" when prompted.

Step 3: Create project folder

Open PowerShell or Command Prompt:

mkdir C:\openclaw
cd C:\openclaw

Step 4: Create docker-compose.yml

Create a file named docker-compose.yml with this content:

version: '3.8'
services:
  openclaw:
    image: openclaw/openclaw:latest
    container_name: openclaw
    restart: unless-stopped
    ports:
      - "8080:8080"
    environment:
      - ANTHROPIC_API_KEY=your-key-here
      - GATEWAY_TOKEN=your-token-here
    volumes:
      - ./data:/app/data

Step 5: Start OpenClaw

docker compose up -d

Auto-Start on Boot

WSL2 Method

Create a startup script to automatically start Docker and OpenClaw:

  1. 1. Press Win+R, type shell:startup
  2. 2. Create a file start-openclaw.vbs with:
Set objShell = CreateObject("WScript.Shell")
objShell.Run "wsl -d Ubuntu -u root service docker start", 0, True
objShell.Run "wsl -d Ubuntu docker compose -f /home/YOUR_USER/openclaw/docker-compose.yml up -d", 0, False

Replace YOUR_USER with your WSL username.

GPU Acceleration (Optional)

If you have an NVIDIA GPU, you can enable GPU acceleration for faster local AI inference:

  1. 1. Install latest NVIDIA drivers (version 470+)
  2. 2. Enable NVIDIA Container Toolkit in WSL2
  3. 3. Add deploy: resources: reservations: devices: to docker-compose.yml
NVIDIA CUDA on WSL Documentation →

Need Help?

Don't want to set it up yourself? We can help.