Paper2Poster/docker-compose.yml at main · TypesetIO/Paper2Poster · 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
services:
  paper2poster:
    build:
      context: .
      dockerfile: Dockerfile
    container_name: paper2poster-app
    environment:
      - PYTHONPATH=/app
      - OPENAI_API_KEY=${OPENAI_API_KEY:-}
      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
      - GOOGLE_AI_API_KEY=${GOOGLE_AI_API_KEY:-}
      - SKIP_MODEL_DOWNLOAD=${SKIP_MODEL_DOWNLOAD:-false}
    volumes:
      # Mount input/output directories
      - ./input_papers:/app/input_papers
      - ./output_posters:/app/output_posters
      - ./tmp:/app/tmp
      - ./contents:/app/contents
      - ./tree_splits:/app/tree_splits
      - ./images_and_tables:/app/images_and_tables
      - ./generated_posters:/app/generated_posters
      # Mount for model cache (required for OCR/parsing models)
      - ./model_cache:/app/model_cache
      # Mount for Docling cache
      - ./model_cache/docling:/root/.cache/docling
      # Mount environment file
      - ./.env:/app/.env:ro
    working_dir: /app
    ports:
      - "6025:6025"
    restart: unless-stopped
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:6025/health"]
      interval: 30s
      timeout: 10s
      retries: 3
    profiles:
      - default

  # Optional: vLLM service for local model deployment
  vllm-server:
    image: vllm/vllm-openai:latest
    container_name: vllm-server
    ports:
      - "8000:8000"
    environment:
      - MODEL_NAME=Qwen/Qwen2.5-7B-Instruct
    command: >
      --model Qwen/Qwen2.5-7B-Instruct
      --host 0.0.0.0
      --port 8000
      --served-model-name vllm_qwen
    volumes:
      - ./model_cache:/root/.cache/huggingface
    profiles:
      - vllm
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

  # Optional: vLLM VL service for vision-language models
  vllm-vl-server:
    image: vllm/vllm-openai:latest
    container_name: vllm-vl-server
    ports:
      - "8001:8001"
    environment:
      - MODEL_NAME=Qwen/Qwen2.5-VL-7B-Instruct
    command: >
      --model Qwen/Qwen2.5-VL-7B-Instruct
      --host 0.0.0.0
      --port 8001
      --served-model-name vllm_qwen_vl
    volumes:
      - ./model_cache:/root/.cache/huggingface
    profiles:
      - vllm
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]

volumes:
  model_cache:
    driver: local

networks:
  default:
    name: paper2poster-network