services:
  ollama:
    image: 'ollama/ollama:latest'
    volumes:
        - ollama_data:/root/.ollama
    ports:
      - '11434:11434'
    mem_limit: 4g
    memswap_limit: 16g
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]
    # healthcheck:
    #   test: ["CMD", "curl", "-f", "http://localhost:11434/api/tags"]
    #   interval: 10s
    #   timeout: 5s
    #   retries: 3
    #   start_period: 20s
  
#  ollama-pull:
#    image: 'ollama/ollama:latest'
    # depends_on:
    #   ollama:
    #     condition: service_healthy
#    volumes:
#      - ollama_data:/root/.ollama
#    environment:
#      - OLLAMA_HOST=http://ollama:11434
#    command: ollama pull llama3.2:latest nomic-embed-text
#    restart: "no"
  
  pgvector:
    image: 'pgvector/pgvector:pg16'
    environment:
      - 'POSTGRES_DB=gcp_docs'
      - 'POSTGRES_PASSWORD=password'
      - 'POSTGRES_USER=admin'
    ports:
      - '15432:5432'
    volumes:
      - ./schema.sql:/docker-entrypoint-initdb.d/schema.sql
      - ./pgdata:/var/lib/postgresql/data

volumes:
  pgdata:
  ollama_data: