lightspeed-stack/tests/e2e-prow/rhoai/configs/run.yaml at main · lightspeed-core/lightspeed-stack · 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
version: 2
distro_name: starter

apis:
- responses
- batches
- files
- inference
- tool_runtime
- conversations
- vector_io

providers:
  inference:
  - provider_id: vllm
    provider_type: remote::vllm
    config:
      base_url: ${env.KSVC_URL}/v1/
      api_token: ${env.VLLM_API_KEY}
      tls_verify: false
      max_tokens: 512
  - config: {}
    provider_id: sentence-transformers
    provider_type: inline::sentence-transformers
  files:
  - config:
      metadata_store:
        table_name: files_metadata
        backend: sql_default
      storage_dir: /opt/app-root/src/.llama/storage/files
    provider_id: meta-reference-files
    provider_type: inline::localfs
  tool_runtime:
  - config: {} # Enable the RAG tool
    provider_id: file-search
    provider_type: inline::file-search
  - config: {} # Enable MCP (Model Context Protocol) support
    provider_id: model-context-protocol
    provider_type: remote::model-context-protocol
  vector_io:
  - config: # Define the storage backend for RAG
      persistence:
        namespace: vector_io::faiss
        backend: kv_rag
    provider_id: faiss
    provider_type: inline::faiss
  responses:
  - config:
      persistence:
        responses:
          table_name: agents_responses
          backend: sql_default
    provider_id: builtin
    provider_type: inline::builtin
  batches:
  - config:
      sqlstore:
        table_name: batches
        backend: sql_default
    provider_id: reference
    provider_type: inline::reference
server:
  port: 8321
storage:
  backends:
    kv_default:  # Single database for registry AND RAG data
      type: kv_sqlite
      db_path: /opt/app-root/src/.llama/storage/rag/kv_store.db
    kv_rag:
      type: kv_sqlite
      db_path: /opt/app-root/src/.llama/storage/rag/kv_store.db
    sql_default:
      type: sql_sqlite
      db_path: ${env.SQL_STORE_PATH:=/opt/app-root/src/.llama/storage/sql_store.db}
  stores:
    metadata:
      namespace: registry
      backend: kv_default
    inference:
      table_name: inference_store
      backend: sql_default
      max_write_queue_size: 10000
      num_writers: 4
    conversations:
      table_name: openai_conversations
      backend: sql_default
    prompts:
      table_name: prompts
      backend: sql_default
    connectors:
      table_name: connectors
      backend: sql_default
registered_resources:
  models:
  - model_id: meta-llama/Llama-3.1-8B-Instruct
    provider_id: vllm
    model_type: llm
    provider_model_id: null
  vector_stores: []
vector_stores:
  default_provider_id: faiss
  default_embedding_model: # Define the default embedding model for RAG
    provider_id: sentence-transformers
    model_id: all-mpnet-base-v2