version: "3.8"

services:
  ollama:
    image: ollama/ollama:latest
    container_name: ollama
    restart: unless-stopped
    ports:
      - "11434:11434" # akses dari localhost / IDE
    volumes:
      - ./ollama/models:/root/.ollama
    environment:
      - OLLAMA_HOST=0.0.0.0
      - OLLAMA_KEEP_ALIVE=15m
      - OLLAMA_NUM_PARALLEL=2 # hemat RAM & CPU
      - OLLAMA_MAX_LOADED_MODELS=1
    networks:
      - llm_network
  # ollama-init:
  #   image: ollama/ollama:latest
  #   container_name: ollama_init
  #   depends_on:
  #     - ollama
  #   entrypoint: >
  #     /bin/sh -c " sleep 10 && ollama pull qwen2.5:7b && echo 'Qwen2.5-7B model pulled' "
  #   networks:
  #     - llm_network

networks:
  llm_network:
    driver: bridge