System Requirements
- OS: Linux, macOS, or Windows 10+
- Python: 3.8 or higher
- Node.js: 16.0 or higher
- Memory: 4GB RAM minimum
- Storage: 2GB free space
- Network: Internet connection for AI API calls
Prerequisites
Before installing DeepWiki-Open, ensure you have:1
Python Installation
Verify Python installation:
python --version
should show 3.8+2
Node.js Installation
Verify Node.js installation:
node --version
should show 16.0+3
Git Installation
Verify Git installation:
git --version
Installation Methods
Choose the installation method that best fits your needs:Docker Installation
Docker provides the easiest and most consistent setup experience.Prerequisites for Docker
1
Install Docker
Download and install Docker Desktop for Mac
2
Install Docker Compose
Docker Compose is typically included with Docker Desktop. For Linux:
Docker Setup
1
Clone Repository
2
Configure Environment
3
Start Services
This starts both backend (port 8001) and frontend (port 3000) services.
Verification
After installation, verify that everything is working correctly:1
Check Backend Health
2
Check Frontend
Visit http://localhost:3000 in your browser.You should see:
- DeepWiki-Open interface
- Repository URL input field
- Model selection dropdown
- Generate Wiki button
3
Test Full Workflow
- Enter a small public repository (e.g.,
https://github.com/octocat/Hello-World
) - Select an AI model provider
- Click “Generate Wiki”
- Verify wiki generation completes successfully
Optional Dependencies
Ollama (Local AI Models)
Ollama (Local AI Models)
To run AI models locally using Ollama:Update your
.env
:Redis (Caching)
Redis (Caching)
For improved performance in production:Add to
.env
:Troubleshooting Installation
Python Installation Issues
Python Installation Issues
Problem:
python: command not found
Solutions:Node.js Installation Issues
Node.js Installation Issues
Problem:
npm ERR! EACCES: permission denied
Solutions:Docker Issues
Docker Issues
Problem:
Cannot connect to Docker daemon
Solutions:Port Conflicts
Port Conflicts
Problem:
Port already in use
Solutions: