multimind-sdk/docker-compose.yml at develop · multimindlab/multimind-sdk · 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
services:
  gateway:
    build:
      context: .
      dockerfile: Dockerfile
    image: ghcr.io/multimindlab/multimind-sdk:latest
    ports:
      - "8000:8000"
    env_file:
      - path: .env
        required: false
    environment:
      # Provider API keys — all optional; configure only the ones you use
      - OPENAI_API_KEY=${OPENAI_API_KEY:-}
      - ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY:-}
      - GROQ_API_KEY=${GROQ_API_KEY:-}
      - MISTRAL_API_KEY=${MISTRAL_API_KEY:-}
      - GEMINI_API_KEY=${GEMINI_API_KEY:-}
      - DEEPSEEK_API_KEY=${DEEPSEEK_API_KEY:-}
      - HUGGINGFACE_API_KEY=${HUGGINGFACE_API_KEY:-}
      # Local models via the optional ollama service (profile: local)
      - OLLAMA_API_BASE=${OLLAMA_API_BASE:-http://ollama:11434}
      - OLLAMA_MODEL_NAME=${OLLAMA_MODEL_NAME:-mistral}
      - DEFAULT_MODEL=${DEFAULT_MODEL:-openai}
      - LOG_LEVEL=${LOG_LEVEL:-INFO}
      - MULTIMIND_PROXY_AUDIT_LOG=/data/audit.jsonl
      - MULTIMIND_PROXY_COSTS_LOG=/data/costs.jsonl
      - MULTIMIND_DASHBOARD_AUDIT_LOG=/data/audit.jsonl
      - MULTIMIND_DASHBOARD_COSTS_LOG=/data/costs.jsonl
      - MULTIMIND_DASHBOARD_GUARDRAILS=/data/guardrails.json
      - MULTIMIND_CORS_ORIGINS=http://localhost:8501,http://127.0.0.1:8501
    volumes:
      - multimind_sessions:/app
      - multimind_governance:/data
    healthcheck:
      test: ["CMD", "python", "-c", "import urllib.request; urllib.request.urlopen('http://127.0.0.1:8000/health', timeout=4)"]
      interval: 30s
      timeout: 5s
      retries: 3
      start_period: 15s
    restart: unless-stopped

  dashboard:
    image: ghcr.io/multimindlab/multimind-sdk:latest
    depends_on:
      - gateway
    command: ["multimind", "dashboard", "--host", "0.0.0.0", "--port", "8501"]
    ports:
      - "8501:8501"
    environment:
      - MULTIMIND_API_BASE=http://localhost:8000
      - MULTIMIND_DASHBOARD_HOST=0.0.0.0
      - MULTIMIND_DASHBOARD_AUDIT_LOG=/data/audit.jsonl
      - MULTIMIND_DASHBOARD_COSTS_LOG=/data/costs.jsonl
      - MULTIMIND_DASHBOARD_GUARDRAILS=/data/guardrails.json
    volumes:
      - multimind_governance:/data
    restart: unless-stopped

  # Optional: local models, no API keys needed.
  # Start with: docker compose --profile local up
  ollama:
    image: ollama/ollama:latest
    profiles: ["local"]
    ports:
      - "11434:11434"
    volumes:
      - ollama_models:/root/.ollama
    healthcheck:
      test: ["CMD", "ollama", "list"]
      interval: 30s
      timeout: 10s
      retries: 3
      start_period: 20s
    restart: unless-stopped

  # Optional: Chroma vector store for RAG workloads.
  # Start with: docker compose --profile vector-store up
  chroma:
    image: chromadb/chroma:latest
    profiles: ["vector-store"]
    ports:
      - "8001:8000"
    volumes:
      - chroma_data:/chroma/chroma
    environment:
      - ANONYMIZED_TELEMETRY=false
    restart: unless-stopped

volumes:
  multimind_sessions:
  multimind_governance:
  ollama_models:
  chroma_data: