LLM_CodeTranslation/docker-compose.yml at main · faetschi/LLM_CodeTranslation · 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
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
networks:
  rabbitmq_network:
    name: rabbitmq_network
    driver: bridge

services:
  ollama:
    image: ollama/ollama
    container_name: ollama
    networks:
      - rabbitmq_network
    ports:
      - "11434:11434"
    volumes:
      - ./docker/ollama:/root/.ollama
    environment:
    - OLLAMA_KEEP_ALIVE=-1        # 10m = unload if unused for 10 minutes # -1 keeps the model in memory indefinitely (needs high VRAM!!!)
    - OLLAMA_MODEL=${LLM_MODEL}   # Comma-separated model list (your .env)
    # - OLLAMA_DEBUG=1              # verbose logs
    - OLLAMA_REQUEST_TIMEOUT=120  # increase timeouts
    - OLLAMA_RUN_TIMEOUT=300
    - OLLAMA_LOAD_TIMEOUT=30m
    - NVIDIA_VISIBLE_DEVICES=all  # <-- Expose all GPUs
    - NVIDIA_DRIVER_CAPABILITIES=compute,utility  # <-- Enable full CUDA performance
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all # <-- Use ALL GPUs (change to 1 if you want just 1 GPU)
              capabilities: [gpu]
    restart: unless-stopped # always

  rabbitmq:
    image: rabbitmq:management
    container_name: rabbitmq
    networks:
      - rabbitmq_network
    ports:
      - "5672:5672"
      - "15672:15672"
    volumes:
      - rabbitmq_data:/var/lib/rabbitmq
    restart: always

  fastapi:
    build:
      context: ./docker
      dockerfile: fastapi/Dockerfile
    container_name: fastapi_service
    networks:
      - rabbitmq_network
    ports:
      - "8000:8000"
    depends_on:
      - rabbitmq
      - ollama
    volumes:
      - shared_volume:/fastapi/uploads
      - ./docker/translation_worker/translated:/translation_worker/translated
      - ./docker/translation_worker/temp:/translation_worker/temp
    restart: always

  translation_worker:
    build:
      context: ./docker
      dockerfile: translation_worker/Dockerfile
    container_name: translation_worker
    networks:
      - rabbitmq_network
    depends_on:
      - rabbitmq
      - ollama
    volumes:
      - shared_volume:/fastapi/uploads
      - ./docker/translation_worker/temp:/translation_worker/temp
      - ./docker/translation_worker/profiles:/translation_worker/profiles
      - ./output:/output
    restart: always
    env_file:
      - .env
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              capabilities: [gpu]

  # test_worker:
  #   build:
  #     context: ./docker
  #     dockerfile: test_worker/Dockerfile
  #   container_name: test_worker
  #   networks:
  #     - rabbitmq_network
  #   depends_on:
  #     - rabbitmq
  #     - ollama
  #   volumes:
  #     - shared_volume:/fastapi/uploads
  #     - ./docker/translation_worker/translated:/translation_worker/translated
  #     - ./docker/tests:/tests
  #     - ./output:/output
  #   restart: always
  #   env_file:
  #     - .env
  #   deploy:
  #     resources:
  #       reservations:
  #         devices:
  #           - driver: nvidia
  #             capabilities: [gpu]

  model-preloader:
    build:
      context: ./docker
      dockerfile: model_loader/Dockerfile
    container_name: model_preloader
    depends_on:
      - ollama
    networks:
      - rabbitmq_network
    env_file:
      - .env
    environment:
      - OLLAMA_URL=http://ollama:11434
      - OLLAMA_URL=http://ollama:11434

volumes:
  shared_volume:
  rabbitmq_data: