mirror of
https://github.com/open-webui/open-webui.git
synced 2026-07-12 02:54:03 -05:00
[GH-ISSUE #13228] feat: Other LLMs Provider's API native support. #120165
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#120165
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @Mike-7777777 on GitHub (Apr 25, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/13228
Check Existing Issues
Problem Description
Now openwebui only support OpenAI style API endpoint. As the market is growing, I believe if we can support the other prividers API will be very helpful for new users. And I think make it navie support is better than run a LiteLLM or something like that.
Desired Solution you'd like
User can select Gemini's, Claude's, Grok's, DeepSeek's, API sytle
Alternatives Considered
No response
Additional Context
No response
@Classic298 commented on GitHub (Apr 25, 2025):
As written in 100s of issues and discussions: Open WebUI will not support any other APIs besides OpenAI compatible API's.
Why?
Because there are 200+ LLM providers. It is IMPOSSIBLE to 1) add all of them and 2) maintain all the code.
Also with so many different providers, it is way too difficult to update any changes in a short timeframe (let alone impossible).
Instead, you should use providers who do support OpenAI compatible APIs like OpenAI, OpenRouter, and even Google Vertex offers a (beta) version of an OpenAI compatible endpoint. Or you use a middleware like LiteLLM, which was built to integrate 100+ LLM providers into a single software which then provides a unified API endpoint (OpenAI compatible) which you can connect to Open WebUI.
Also, if you start integrating other providers, users will ask why X gets integrated and Y is not integrated yet or has been assigned a lower priority (besides the other mentioned issues: too many, hard to maintain, impossible to keep up).
Ask your provider to provide an OpenAI compatible endpoint or use OpenRouter or middleware like LiteLLM.
I'd advise you to close this issue, otherwise the Maintainers will have to do it.
And please search for existing issues and discussions next time, this has been addressed easily over 100 times.