version: '3.8'

services:
  chatbot-ml:
    build:
      context: .
      args:
        # For GPU support, uncomment the next line:
        # BASE_IMAGE: pytorch/pytorch:2.2.0-cuda12.1-cudnn8-runtime
        BASE_IMAGE: python:3.11-slim
    ports:
      - "8000:8000"
    environment:
      - MONGODB_URI=mongodb://localhost:27017/wanderlust
      - USE_LLM_GENERATOR=false
      - BASE_MODEL_NAME=Qwen/Qwen2.5-1.5B-Instruct
      - LLM_QUANTIZATION=4bit
      - LLM_TIMEOUT_SECONDS=10
    volumes:
      - ./saved_models:/app/saved_models
    restart: unless-stopped
    # For GPU support, uncomment the following:
    # deploy:
    #   resources:
    #     reservations:
    #       devices:
    #         - driver: nvidia
    #           count: 1
    #           capabilities: [gpu]