LS-core-test/run-llama-stack.yaml at master · rhdhorchestrator/LS-core-test · 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
version: '3'
image_name: ollama-llama-stack-config
apis:
  - agents
  - inference
  - safety
  - telemetry
  - tool_runtime
  - vector_io
logging:
  level: DEBUG  # Set root logger to DEBUG
  category_levels:
    llama_stack: DEBUG  # Enable DEBUG for all llama_stack modules
    llama_stack.providers.remote.inference.vllm: DEBUG
    llama_stack.providers.inline.agents.meta_reference: DEBUG
    llama_stack.providers.inline.agents.meta_reference.agent_instance: DEBUG
    llama_stack.providers.inline.vector_io.faiss: DEBUG
    llama_stack.providers.inline.telemetry.meta_reference: DEBUG
    llama_stack.core: DEBUG
    llama_stack.apis: DEBUG
    uvicorn: DEBUG
    uvicorn.access: INFO  # Keep HTTP requests at INFO to reduce noise
    fastapi: DEBUG

providers:
  vector_io:
    - config:
        kvstore:
          db_path: /tmp/faiss_store.db
          type: sqlite
      provider_id: faiss
      provider_type: inline::faiss

  agents:
  - config:
      persistence_store:
        db_path: /tmp/agents_store.db
        namespace: null
        type: sqlite
      responses_store:
        db_path: /tmp/responses_store.db
        type: sqlite
    provider_id: meta-reference
    provider_type: inline::meta-reference


  inference:
    - provider_id: vllm-inference
      provider_type: remote::vllm
      config:
        url: ${env.VLLM_URL:=http://localhost:8000/v1}
        max_tokens: ${env.VLLM_MAX_TOKENS:=4096}
        api_token: ${env.VLLM_API_TOKEN:=fake}
        tls_verify: ${env.VLLM_TLS_VERIFY:=false}

    - provider_id: google-vertex
      provider_type: remote::vertexai
      config:
        project: ${env.VERTEXAI_PROJECT}
        region: ${env.VERTEXAI_REGION:=us-east5}

  tool_runtime:
    - provider_id: model-context-protocol
      provider_type: remote::model-context-protocol
      config: {}
      module: null

  telemetry:
    - config:
        service_name: 'llama-stack'
        sinks: console,sqlite
        sqlite_db_path: /tmp/trace_store.db
      provider_id: meta-reference
      provider_type: inline::meta-reference

metadata_store:
  type: sqlite
  db_path: /tmp/registry.db
  namespace: null

inference_store:
  type: sqlite
  db_path: /tmp/inference_store.db