Repository navigation
[FEATURE]: Use Responses api for providers that support it #10601
Description
Feature hasn't been suggested before.
- I have verified this feature I'm about to request hasn't been suggested before.
Describe the enhancement you want to request
Now that openresponses.org is out and adopted by many providers (including Huggingface, LM Studio, OpenRouter), I think it would be beneficial if these providers used the v1/responses api rather than the chat/completions API. From my testing, it seems as though its as simple as setting the npm field to be @ai-sdk/openai rather than @ai-sdk/openai-compatible and it will default to responses as documented here. Also, might be helpful to change the docs for setting up LM Studio to use the openai provider rather than openai-compatible. Not sure if this is a change that needs to take place here or on the models.dev repo but I thought I'd put it here.
Reacted by Simão Gonçalves, Andrew A, Michael Heyns, Dyllon Gunawardhana and Enty
Metadata
Metadata
Assignees
Labels
No labels
Activity
github-actions commentedon Jan 26, 2026
This issue might be a duplicate of existing issues. Please check:
/v1/responsesfor xAI Providers #3890: Requests support for /v1/responses endpoint for xAI providersIssues #9628 and #7793 appear to be the strongest duplicates as they directly address the same feature request and underlying problem.
Feel free to ignore if this doesn't address your specific case.
openSourcerer9000 commentedon Feb 28, 2026
Any update on this? LM studio locally is rough to set up first of all (your LM Studio (local) routes to some unknown API), but it's impossible to use because it doesn't have persistent context.
github-actions commentedon Jul 6, 2026
To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.