services:
  glados-tts-cpu:
    build:
      context: .
      dockerfile: Dockerfile
    image: glados-tts-wyoming:cpu
    container_name: glados-tts-cpu
    ports:
      - "10201:10200"
    volumes:
      - glados_model_cache:/data
    environment:
      - MODEL_DIR=/data
      - URI=tcp://0.0.0.0:10200
      - DEVICE=cpu
    restart: unless-stopped

  glados-tts-cuda:
    build:
      context: .
      dockerfile: Dockerfile.cuda
    image: glados-tts-wyoming:cuda
    container_name: glados-tts-cuda
    ports:
      - "10200:10200"
    volumes:
      - glados_model_cache:/data
    environment:
      - MODEL_DIR=/data
      - URI=tcp://0.0.0.0:10200
      - DEVICE=cuda
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    restart: unless-stopped

  glados-tts-rocm:
    build:
      context: .
      dockerfile: Dockerfile.rocm
    image: glados-tts-wyoming:rocm
    container_name: glados-tts-rocm
    ports:
      - "10202:10200"
    volumes:
      - glados_model_cache:/data
    environment:
      - MODEL_DIR=/data
      - URI=tcp://0.0.0.0:10200
      - DEVICE=rocm
    devices:
      - /dev/kfd
      - /dev/dri
    restart: unless-stopped

volumes:
  glados_model_cache: