modelsynic/docker-compose.vllm.yml at main · devhttps/modelsynic · 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
version: '3.8'

services:
  modelsync-vllm:
    build:
      context: .
      dockerfile: Dockerfile.vllm
    ports:
      - "8001:8001"  # vLLM API
      - "8000:8000"  # ModelSync API
      - "8080:8080"  # Web interface
    volumes:
      - .:/app
      - modelsync-data:/app/.modelsync
      - vllm-models:/app/models
    environment:
      - CUDA_VISIBLE_DEVICES=0
      - PYTHONPATH=/app
      - VLLM_USE_MODELSCOPE=False
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]
    restart: unless-stopped
    command: ["python3", "-m", "modelsync.llm.vllm_api"]

  modelsync-api:
    build:
      context: .
      dockerfile: Dockerfile
    ports:
      - "8002:8000"  # ModelSync API standalone
    volumes:
      - .:/app
      - modelsync-data:/app/.modelsync
    environment:
      - PYTHONPATH=/app
    restart: unless-stopped
    command: ["python3", "-m", "modelsync.api.main"]

  modelsync-web:
    build:
      context: .
      dockerfile: Dockerfile
    ports:
      - "8081:8080"  # Web interface standalone
    volumes:
      - .:/app
      - modelsync-data:/app/.modelsync
    environment:
      - PYTHONPATH=/app
    restart: unless-stopped
    command: ["python3", "-m", "modelsync.web.app"]

volumes:
  modelsync-data:
  vllm-models: