kayx23 opened a new pull request, #13731:
URL: https://github.com/apache/apisix/pull/13731

   ### Description
   
   Document how `ai-proxy` and `ai-proxy-multi` distinguish OpenAI Chat 
Completions, Responses API, and Embeddings requests.
   
   - add the exact protocol detection rules and explain the `input` overlap 
between Responses and Embeddings
   - add minimal Responses API route and request examples
   - clarify that the `openai` provider selects the endpoint for the detected 
protocol
   - remove redundant OpenAI Embeddings endpoint overrides from the examples
   - update both English and Chinese plugin pages
   
   The Responses and Embeddings formats both use `input`, but only Responses 
also requires the request URI to end in `/v1/responses`. This distinction and 
the provider endpoint selection were not previously explained, which made 
custom Route URIs and Embeddings configuration easy to misunderstand.
   
   #### Which issue(s) this PR fixes:
   
   N/A
   
   ### Validation
   
   - verified the documented detection order and OpenAI endpoint capabilities 
against the implementation and existing protocol tests
   - `markdownlint` passed for all four changed pages
   - `utils/fix-zh-doc-segment.py` passed
   - `git diff --check` passed
   
   ### Checklist
   
   - [x] I have explained the need for this PR and the problem it solves
   - [x] I have explained the changes or the new features added to this PR
   - [ ] I have added tests corresponding to this change (documentation-only; 
existing protocol tests cover this behavior)
   - [x] I have updated the documentation to reflect this change
   - [x] I have verified that this change is backward compatible


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to