services:
  llamalink:
    build:
      context: .
      dockerfile: deploy/Dockerfile
    container_name: llamalink-api
    ports:
      - "8000:8000"
    environment:
      - LLAMALINK_ENV=production
      - DATABASE_URL=sqlite+aiosqlite:///./data/llamalink.db
      - MANAGE_LLAMA_SERVER=true
      - LLAMA_SERVER_HOST=127.0.0.1
      - LLAMA_SERVER_PORT=8080
      - LLAMA_SERVER_BIN=/usr/local/bin/llama-server
      - LLAMA_SERVER_STARTUP_TIMEOUT=120
      - MODEL_SWAP_COOLDOWN=2
      - LLAMA_SERVER_STOP_TIMEOUT=10
      - RATE_LIMIT_PER_MINUTE=60
      - RATE_LIMIT_STORAGE=memory
      - LOG_LEVEL=info
      - LOG_FORMAT=json
      - ADMIN_TOKEN=${ADMIN_TOKEN}
    volumes:
      - llamalink-data:/app/data
      - ./models:/models:ro
    restart: unless-stopped
    networks:
      - llamalink-net
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 10s

  llama-runner:
    image: ghcr.io/ggml-org/llama.cpp:server
    container_name: llama-runner
    environment:
      - CUDA_VISIBLE_DEVICES=0
    volumes:
      - ./models:/models:ro
    restart: unless-stopped
    networks:
      - llamalink-net
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/health"]
      interval: 30s
      timeout: 10s
      retries: 3
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

  nginx:
    image: nginx:1.27-alpine
    container_name: llamalink-nginx
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - ./deploy/nginx.conf:/etc/nginx/nginx.conf:ro
    depends_on:
      - llamalink
    restart: unless-stopped
    networks:
      - llamalink-net

volumes:
  llamalink-data:

networks:
  llamalink-net:
    driver: bridge