services:
  llm-engine:
    build:
      context: .
      dockerfile: Dockerfile
      args:
        MODEL_ID: TinyLlama/TinyLlama-1.1B-Chat-v1.0
    image: llm-serving-engine:latest
    container_name: llm-engine
    ports:
      - "8000:8000"
    environment:
      # Override any of these to change runtime behaviour without rebuilding
      - HF_HOME=/app/.hf_cache
    # Mount a volume so HuggingFace weights are cached across rebuilds
    volumes:
      - hf-model-cache:/app/.hf_cache
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health')"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 120s   # give the model time to load

volumes:
  hf-model-cache:
    driver: local