local-ai-server/docker-compose.yml at main · sulinivan/local-ai-server · 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
services:
  llama-server:
    image: ggml-org/llama.cpp:full-cuda
    restart: unless-stopped
    ipc: host
    runtime: nvidia
    environment:
      HF_HUB_OFFLINE: "1"
      TRANSFORMERS_OFFLINE: "1"
      NVIDIA_VISIBLE_DEVICES: all
    volumes:
      - ./models:/models:ro,Z
      - ./presets.ini:/presets/presets.ini:ro,Z
    command:
      - "--server"
      - "--host"
      - "0.0.0.0"
      - "--port"
      - "8000"
      - "--api-key"
      - "default-key-change-me"
      - "--models-dir"
      - "/models"
      - "--models-preset"
      - "/presets/presets.ini"
      - "--cache-ram"
      - "-1"
      - "--cache-reuse"
      - "256"
      - "--parallel"
      - "2"
    healthcheck:
      test: ["CMD", "curl", "-sf", "http://localhost:8000/models"]
      interval: 15s
      timeout: 10s
      retries: 10
      start_period: 90s
    networks:
      - backend

  open-webui:
    image: open-webui/open-webui:cuda
    restart: unless-stopped
    ipc: host
    runtime: nvidia
    depends_on:
      llama-server:
        condition: service_healthy
    environment:
      OPENAI_API_BASE_URL: http://llama-server:8000/v1
      OPENAI_API_KEY: default-key-change-me
      WEBUI_PROXY_ENABLED: "true"
      WEBUI_PROXY_ALLOWED_ORIGINS: "http://192.168.1.100,https://ai-server.local,http://localhost,https://localhost"
      OFFLINE_MODE: true
      HF_HUB_OFFLINE: "1"
      NVIDIA_VISIBLE_DEVICES: all
    volumes:
      - ./data/webui:/app/backend/data:Z
    healthcheck:
      test: ["CMD", "curl", "-sf", "http://localhost:8080/"]
      interval: 15s
      timeout: 10s
      retries: 10
      start_period: 90s
    networks:
      - backend

  nginx:
    image: nginx:1.27-alpine
    restart: unless-stopped
    depends_on:
      - open-webui
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - ./data/nginx/nginx.conf:/etc/nginx/nginx.conf:ro,Z
      - ./data/nginx/certs:/etc/nginx/certs:ro,Z
    networks:
      - backend
      - public

networks:
  public:
    driver: bridge
  backend:
    driver: bridge
    internal: true