ai_platform/docker-compose.yml at master · RomanShushakov/ai_platform · 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
services:
  ollama:
    image: ollama/ollama:0.21.0
    container_name: ai_platform_ollama
    ports:
      - "11434:11434"
    volumes:
      - ollama_data:/root/.ollama
    healthcheck:
      test: [ "CMD", "ollama", "list" ]
      interval: 10s
      timeout: 5s
      retries: 10

  host:
    build:
      context: .
      dockerfile: host/dockerfile
    container_name: ai_platform_host
    depends_on:
      - ollama
    environment:
      HOST_PORT: "${HOST_PORT:-3000}"

      TOOL_PROVIDER: "${TOOL_PROVIDER:-mcp}" # options: http, mcp
      TOOLS_BASE_URL: "${TOOLS_BASE_URL:-http://tools-server:3001}"
      MCP_TOOLS_BINARY: "${MCP_TOOLS_BINARY:-/usr/local/bin/tools-server}"

      LLM_BACKEND: "${LLM_BACKEND:-ollama}"
      LLM_BASE_URL: "${LLM_BASE_URL:-http://ollama:11434}"
      LLM_CHAT_PATH: "${LLM_CHAT_PATH:-/api/chat}"
      LLM_MODEL: "${LLM_MODEL:-llama3}"

      RETRIEVAL_BACKEND: "${RETRIEVAL_BACKEND:-embeddings_local}" # options: noop, inmemory_markdown, embeddings_local
      RETRIEVAL_TOP_K: "${RETRIEVAL_TOP_K:-4}"
      KNOWLEDGE_BASE_PATH: "${KNOWLEDGE_BASE_PATH:-/app/knowledge_base}"

      MAX_LLM_STEPS: "${MAX_LLM_STEPS:-4}"

      RAG_ARTIFACTS_PATH: "${RAG_ARTIFACTS_PATH:-/app/artifacts/rag}"
      EMBEDDING_BACKEND: "${EMBEDDING_BACKEND:-ollama}"
      EMBEDDING_BASE_URL: "${EMBEDDING_BASE_URL:-http://ollama:11434}"
      EMBEDDING_MODEL: "${EMBEDDING_MODEL:-nomic-embed-text}"

      RETRIEVAL_MIN_SCORE: "${RETRIEVAL_MIN_SCORE:-0.45}"
      RETRIEVAL_RELATIVE_RATIO: "${RETRIEVAL_RELATIVE_RATIO:-0.95}"
      RETRIEVAL_USE_THRESHOLD: "${RETRIEVAL_USE_THRESHOLD:-0.12}"

      RUST_LOG: "${RUST_LOG:-host=debug,llm_client=debug,tower_http=debug}"

    volumes:
      - ./knowledge_base:/app/knowledge_base
      - ./artifacts:/app/artifacts

    ports:
      - "3000:3000"

  tools-server:
    profiles: [ "http-tools" ]
    build:
      context: .
      dockerfile: tools_server/dockerfile
    container_name: ai_platform_tools_server
    environment:
      TOOLS_TRANSPORT: "http"
      RUST_LOG: "${TOOLS_RUST_LOG:-tools_server=debug,tower_http=debug}"
    ports:
      - "3001:3001"

volumes:
  ollama_data: