scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-09 03:18:44 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 108

Add about managed ids in the doc

Browse source
This commit is contained in:
Sameer Kankute 2026-03-20 16:39:18 +05:30
parent 2634088354
commit ec7c94c61f
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
docs/my-website/docs/tutorials/file_search_responses_api.md
View file

@ -19,7 +19,7 @@ Two paths are covered:
| **Native passthrough** | Provider natively supports `file_search` (OpenAI, Azure) | Decodes unified vector store ID → forwards to provider as-is |
| **Emulated fallback** | Provider doesn't support `file_search` (Anthropic, Bedrock, etc.) | Converts to a function tool → intercepts tool call → runs vector search → synthesizes OpenAI-format output |
In `tools[].vector_store_ids`, LiteLLM accepts both provider-native IDs (e.g. `vs_...`) **and** **managed vector store unified IDs** (URL-safe base64 strings from the proxy managed-vector flow), e.g. `litellm.responses(..., tools=[{"type": "file_search", "vector_store_ids": ["bGl0ZWxsbV9wcm94eT..."]}])`.
## Usage