graphrag-api/docker-compose.yml at main · pierregrothe/graphrag-api · 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
version: "3.8"

services:
    # GraphRAG API - Main Application
    graphrag-api:
        build:
            context: .
            dockerfile: Dockerfile
        container_name: graphrag-api
        ports:
            - "8001:8001"
        environment:
            # Core Settings
            APP_NAME: "GraphRAG API Service"
            DEBUG: "true"
            PORT: 8001

            # LLM Configuration - Ollama for local dev
            LLM_PROVIDER: ollama
            OLLAMA_BASE_URL: http://ollama:11434
            OLLAMA_LLM_MODEL: gemma2:2b # Smaller model for dev
            OLLAMA_EMBEDDING_MODEL: nomic-embed-text

            # Simple SQLite database (no external DB needed!)
            DATABASE_TYPE: sqlite
            DATABASE_PATH: /app/data/graphrag.db

            # Minimal caching (in-memory for simplicity)
            CACHE_TYPE: memory
            CACHE_TTL: 3600

            # Security (simplified for dev)
            JWT_SECRET_KEY: dev-secret-key-change-in-production
            AUTH_ENABLED: false # Disable for local dev

            # Performance (tuned for small scale)
            MAX_WORKERS: 2
            CONNECTION_POOL_SIZE: 5

        volumes:
            # Persist data and GraphRAG files
            - ./data:/app/data
            - ./logs:/app/logs

        healthcheck:
            test: ["CMD", "curl", "-f", "http://localhost:8001/health"]
            interval: 30s
            timeout: 10s
            retries: 3

        depends_on:
            ollama:
                condition: service_healthy

    # Ollama - Local LLM
    ollama:
        image: ollama/ollama:latest
        container_name: ollama
        ports:
            - "11434:11434"
        volumes:
            - ollama_data:/root/.ollama
        healthcheck:
            test: ["CMD", "curl", "-f", "http://localhost:11434/api/tags"]
            interval: 30s
            timeout: 10s
            retries: 5
        environment:
            OLLAMA_KEEP_ALIVE: 24h
            OLLAMA_MAX_LOADED_MODELS: 2

    # Model Loader - Automatically pulls required models
    model-loader:
        image: ollama/ollama:latest
        container_name: model-loader
        depends_on:
            ollama:
                condition: service_healthy
        entrypoint: ["/bin/sh", "-c"]
        command:
            - |
                echo "Waiting for Ollama to be ready..."
                sleep 5
                echo "Pulling Gemma 2B model..."
                ollama pull gemma2:2b || ollama pull gemma:2b
                echo "Pulling embedding model..."
                ollama pull nomic-embed-text
                echo "Models ready!"
        volumes:
            - ollama_data:/root/.ollama

    # Optional: Simple Web UI for testing
    graphrag-ui:
        image: nginx:alpine
        container_name: graphrag-ui
        ports:
            - "3000:80"
        volumes:
            - ./ui:/usr/share/nginx/html:ro
        depends_on:
            - graphrag-api

volumes:
    ollama_data:
        driver: local

networks:
    default:
        name: graphrag-network
        driver: bridge