llm-persistent-instructions/docker-compose.yml at main · mheadd/llm-persistent-instructions · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
services:
  ollama:
    image: ollama/ollama:latest
    container_name: ollama-service
    ports:
      - "11434:11434"
    volumes:
      - ollama_data:/root/.ollama
    environment:
      - OLLAMA_HOST=0.0.0.0
      - OLLAMA_KEEP_ALIVE=60m          # Keep model in memory for 60 minutes
      - OLLAMA_NUM_PARALLEL=2          # Process multiple requests in parallel
      - OLLAMA_MAX_LOADED_MODELS=1     # Keep only 1 model loaded to save memory
    deploy:
      resources:
        limits:
          memory: 8G                   # Maximum memory allocation
        reservations:
          memory: 6G                   # Reserved memory allocation
    shm_size: 2G                       # Shared memory for faster processing
    networks:
      - ai-network
    restart: unless-stopped

  api:
    build: ./api
    container_name: government-ai-api
    ports:
      - "3000:3000"
    volumes:
      - ./api/config:/app/config
      - ./api/.env:/app/.env
    env_file:
      - ./api/.env
    environment:
      - NODE_ENV=production
    depends_on:
      - ollama
    deploy:
      resources:
        limits:
          memory: 1G                   # API doesn't need much memory
        reservations:
          memory: 512M
    networks:
      - ai-network
    restart: unless-stopped

volumes:
  ollama_data:
    driver: local

networks:
  ai-network:
    driver: bridge