docker-compose.yml

services:
  # 
  # Ollama
  # The engine that manages your LLMs
  #
  ollama:
    image: ollama/ollama:latest
    ports:
      - 11434:11434
    volumes:
      - ollama:/root/.ollama
    container_name: ollama
    tty: true
    restart: unless-stopped
    networks:
      - tunnel
# (Uncomment below for Nvidia gpu)
#     deploy:
#       resources:
#         reservations:
#           devices:
#             - driver: nvidia
#               count: all
#               capabilities: [gpu]

  #
  # Open WebUI
  # This container serves a slick ui for you to manage your llms
  # It does no do any ML work under the hood, simply exposes the ollama service via a browser
  #
  open-webui:
    image: ghcr.io/open-webui/open-webui:main
    container_name: open-webui
    volumes:
      - open-webui:/app/backend/data
    depends_on:
      - ollama
    ports:
      - 3000:8080
    environment:
      - 'OLLAMA_BASE_URL=http://ollama:11434'
      - 'WEBUI_SECRET_KEY='
    networks:
      - tunnel
    extra_hosts:
      - host.docker.internal:host-gateway
    restart: unless-stopped
    
  #
  # Cloudflared
  # This container will open a tunnel to cloudflare to expose your site
  # From the tunnel dashboard you will need to:
  # - get a token (which you can set in the environment here)
  # - configure the tunnel to read from http://open-webui:8080 (the other docker container)
  #
  cloudflared:
    image: cloudflare/cloudflared:latest
    container_name: cloudflared
    command: tunnel --no-autoupdate run
    networks:
      - tunnel
    restart: unless-stopped
    environment:
      - TUNNEL_TOKEN=${TUNNEL_TOKEN} # Don't forget to set your tunnel token

volumes:
  ollama: {}
  open-webui: {}
  
networks:
  tunnel: