home_stack/ai_stack/ollama/docker-compose.yml

services:
  ollama:
    build: 
      context: .
      network: host
    image: localhost/ollama-server:latest
    ports:
      - "11434:11434" # Map host port 11434 to container port 11434 (Ollama default)
    volumes:
      - ./models:/models # Mount a local directory for models
      - ollama_data:/usr/share/ollama/.ollama # Persistent storage for Ollama data
    container_name: ollama-server
    environment:
      - OLLAMA_HOST=0.0.0.0
    restart: unless-stopped
    network_mode: slirp4netns
    # Uncomment the following lines if you have NVIDIA GPU support
    # deploy:
    #   resources:
    #     reservations:
    #       devices:
    #         - driver: nvidia
    #           count: all
    #           capabilities: [gpu]

  postgres:
    image: pgvector/pgvector:pg17
    container_name: ollama-postgres
    ports:
      - "54320:5432"
    environment:
      - POSTGRES_PASSWORD=postgres
    volumes:
      - ./postgres-data:/var/lib/postgresql/data
    restart: unless-stopped

  speaches:
    image: ghcr.io/speaches-ai/speaches:latest-cpu
    container_name: speaches-server
    volumes:
      - hf-hub-cache:/home/ubuntu/.cache/huggingface/hub
    ports:
      - "8900:8000"
    # deploy:
    #   resources:
    #     reservations:
    #       # WARN: requires Docker Compose 2.24.2
    #       # https://docs.docker.com/reference/compose-file/merge/#replace-value
    #       devices: !override
    #         - capabilities: ["gpu"]
    #           driver: cdi
    #           device_ids:
    #             - nvidia.com/gpu=all

volumes:
  ollama_data:
  hf-hub-cache: