docker-compose.yml

version: '3.8'

services:
  compact-ai-model:
    build: .
    ports:
      - "8000:8000"
    environment:
      - MODEL_SIZE=small
      - API_HOST=0.0.0.0
      - API_PORT=8000
    volumes:
      - ./checkpoints:/app/checkpoints
      - ./data:/app/data
    deploy:
      resources:
        limits:
          memory: 4G
          cpus: '2.0'
        reservations:
          memory: 2G
          cpus: '1.0'
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
      interval: 30s
      timeout: 10s
      retries: 3
    restart: unless-stopped

  # Optional: Add a reverse proxy
  nginx:
    image: nginx:alpine
    ports:
      - "80:80"
    volumes:
      - ./nginx.conf:/etc/nginx/nginx.conf:ro
    depends_on:
      - compact-ai-model
    profiles:
      - prod