[GH-ISSUE #16445] feat: increase OpenAI API compatibility by redirecting API calls going to '/v1' #104900

Closed
opened 2026-05-18 02:42:25 -05:00 by GiteaMirror · 1 comment
Owner

Originally created by @c0008 on GitHub (Aug 10, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/16445

Check Existing Issues

  • I have searched the existing issues and discussions.

Problem Description

When using Open-WebUI's own API endpoint in other tools there can be compatibility issues with the url format after the base url. In these tools you have to provide a base url, like the string 'http://localhost:8080/api' for open-webui. Then for api requests the tool extends this string, e.g. for the model list it would add '/v1/models', so the resulting url would be 'http://localhost:8080/api/v1/models'. This is working with the OpenAI API, llama.cpp server, Ollama and GPU-Stack but not with Open-WebUI when you want to use it as a unified LLM provider. Open-WebUI would only accept 'http://localhost:8080/api/models'.

Desired Solution you'd like

Handle api calls like '/v1/models' or '/v1/chat/completions' the same as '/models' or '/chat/completions'.

Alternatives Considered

No response

Additional Context

This is not an issue with tools that expect the '/v1' to be part of the base url.

Originally created by @c0008 on GitHub (Aug 10, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/16445 ### Check Existing Issues - [x] I have searched the existing issues and discussions. ### Problem Description When using Open-WebUI's own API endpoint in other tools there can be compatibility issues with the url format after the base url. In these tools you have to provide a base url, like the string 'http://localhost:8080/api' for open-webui. Then for api requests the tool extends this string, e.g. for the model list it would add '/v1/models', so the resulting url would be 'http://localhost:8080/api/v1/models'. This is working with the OpenAI API, llama.cpp server, Ollama and GPU-Stack but not with Open-WebUI when you want to use it as a unified LLM provider. Open-WebUI would only accept 'http://localhost:8080/api/models'. ### Desired Solution you'd like Handle api calls like '/v1/models' or '/v1/chat/completions' the same as '/models' or '/chat/completions'. ### Alternatives Considered _No response_ ### Additional Context This is not an issue with tools that expect the '/v1' to be part of the base url.
Author
Owner

@tjbck commented on GitHub (Aug 10, 2025):

Addressed with 32a022a823

<!-- gh-comment-id:3172881405 --> @tjbck commented on GitHub (Aug 10, 2025): Addressed with 32a022a823eac7d308bc699e319992319fa375ee
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#104900