7df4acdb7d
* docs: stop teaching 0.0.0.0 SurrealDB port exposure in setup snippets Bind port 8000 to 127.0.0.1 in every compose and docker run snippet (README, quick starts, installation, configuration and development docs, and the examples/docker-compose-*.yml files), matching the shipped docker-compose.yml from #1025. Drop the redundant --bind 0.0.0.0:8000 from containerized surreal start commands (it is the in-container default) and add !override to docker-compose.override.yml.example so the opt-in re-publish actually replaces the base port binding instead of colliding with it. Docs that discuss reaching the database from another machine now point at the override example plus a firewall/SSH-tunnel note. Closes #1034 * docs: parameterize creds in manual compose example, note Compose version for !override |
||
|---|---|---|
| .. | ||
| easypanel | ||
| docker-compose-dev.yml | ||
| docker-compose-full-local.yml | ||
| docker-compose-ollama.yml | ||
| docker-compose-single.yml | ||
| docker-compose-speaches.yml | ||
| README.md | ||
Docker Compose Examples
This folder contains different docker-compose.yml configurations for various use cases.
It also includes an easypanel template that can be copied into
the official EasyPanel templates repository and tested in the EasyPanel
templates playground.
📋 Available Examples
docker-compose-full-local.yml - 100% Local AI (No Cloud APIs) 🌟
Use this if: You want complete privacy with zero external API dependencies
Features:
- Ollama: Local LLM and embeddings (mistral, llama, etc.)
- Speaches: Local TTS (text-to-speech) and STT (speech-to-text)
- Everything runs on your machine - nothing sent to cloud
- Perfect for privacy, offline work, or air-gapped environments
Setup:
- Copy to your project folder as
docker-compose.yml - Run:
docker compose up -d - Download models (see file comments for commands)
- Configure all providers in UI (detailed instructions in file)
Requirements:
- Minimum: 8GB RAM, 20GB disk, 4 CPU cores
- Recommended: 16GB+ RAM, NVIDIA GPU (8GB+ VRAM), 50GB disk
Documentation:
docker-compose-speaches.yml - Local Speech Processing
Use this if: You want free TTS/STT but use cloud LLMs
Features:
- Speaches: Local text-to-speech and speech-to-text
- Use with cloud LLM providers (OpenAI, Anthropic, etc.)
- Great for podcast generation without TTS API costs
- Private audio processing
Setup:
- Copy to your project folder as
docker-compose.yml - Run:
docker compose up -d - Download speech models (see file for commands)
- Configure cloud LLM + local Speaches in UI
Documentation:
docker-compose-ollama.yml - Free Local AI with Ollama
Use this if: You want to run AI models locally without API costs
Features:
- Includes Ollama service for local AI models
- No external API keys needed (for LLM and embeddings)
- Full privacy - everything runs on your machine
- Great for testing or privacy-focused deployments
Setup:
- Copy to your project folder as
docker-compose.yml - Run:
docker compose up -d - Pull a model:
docker exec open_notebook-ollama-1 ollama pull mistral - Configure in UI: Settings → API Keys → Add Ollama (URL:
http://ollama:11434)
Recommended models:
- LLM:
mistral,llama3.1,qwen2.5 - Embeddings:
nomic-embed-text,mxbai-embed-large
docker-compose-single.yml - Single Container (Deprecated)
Use this if: You need all services in one container (not recommended)
⚠️ Deprecated: We recommend using the standard multi-container setup (docker-compose.yml in root) for better reliability and easier troubleshooting.
Features:
- Single container includes SurrealDB, API, and Frontend
- Simpler for very constrained environments
- Less flexible for debugging and scaling
docker-compose-dev.yml - Development Setup
Use this if: You're contributing to Open Notebook or developing custom features
Features:
- Hot-reload for code changes
- Separate backend and frontend services
- Build from source instead of using pre-built images
- Includes development tools and debugging
Prerequisites:
- Python 3.11+
- Node.js 18+
- uv (Python package manager)
Setup: See Development Guide
🔄 How to Use These Examples
- Choose the example that fits your use case
- Copy the file to your project folder:
cp examples/docker-compose-ollama.yml docker-compose.yml - Edit the
OPEN_NOTEBOOK_ENCRYPTION_KEYvalue - Run the services:
docker compose up -d
💡 Need a Custom Setup?
You can combine features from multiple examples. Common customizations:
Add Ollama to Standard Setup
Add this to the main docker-compose.yml:
ollama:
image: ollama/ollama:latest
ports:
- "11434:11434"
volumes:
- ollama_models:/root/.ollama
restart: always
volumes:
ollama_models:
Add Reverse Proxy
Add Basic Auth
Add to open_notebook service environment:
- BASIC_AUTH_USERNAME=admin
- BASIC_AUTH_PASSWORD=your-secure-password
📚 Documentation
🆘 Need Help?
- Discord: Join our community
- Issues: GitHub Issues