[GH-ISSUE #10485] Support of multiple llm providers #86648

Closed
opened 2026-05-15 10:58:16 -05:00 by GiteaMirror · 3 comments
Owner

Originally created by @rzafiamy on GitHub (Feb 21, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/10485

Feature Request

Is your feature request related to a problem? Please describe.

The current system only allows for the selection of a single LLM provider, which can be limiting for users who want to compare or switch between multiple providers. This limitation restricts the flexibility and versatility of the platform, making it less adaptable to different user needs and preferences.

Describe the solution you'd like

I would like to see a feature that allows users to select and manage multiple LLM providers simultaneously. This could involve a user interface update to accommodate multiple selections and backend changes to support interactions with multiple providers. For example, users could have a dropdown menu or a list where they can add, remove, or prioritize different LLM providers. This would enable users to compare outputs from different providers, switch between them easily, and leverage the strengths of each provider for different tasks.

Describe alternatives you've considered

The current workarround solution is to use an llm-proxy . But I need to switch manually the provider url each time I want to test it.

Additional context

This feature would significantly enhance the user experience by providing more options and flexibility. It would be particularly useful for users who need to evaluate different LLM providers for specific tasks or who want to ensure they are getting the best possible performance and accuracy. Additionally, supporting multiple providers could attract a wider range of users who have different preferences or requirements for their LLM interactions.

Originally created by @rzafiamy on GitHub (Feb 21, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/10485 # Feature Request **Is your feature request related to a problem? Please describe.** The current system only allows for the selection of a single LLM provider, which can be limiting for users who want to compare or switch between multiple providers. This limitation restricts the flexibility and versatility of the platform, making it less adaptable to different user needs and preferences. **Describe the solution you'd like** I would like to see a feature that allows users to select and manage multiple LLM providers simultaneously. This could involve a user interface update to accommodate multiple selections and backend changes to support interactions with multiple providers. For example, users could have a dropdown menu or a list where they can add, remove, or prioritize different LLM providers. This would enable users to compare outputs from different providers, switch between them easily, and leverage the strengths of each provider for different tasks. **Describe alternatives you've considered** The current workarround solution is to use an llm-proxy . But I need to switch manually the provider url each time I want to test it. **Additional context** This feature would significantly enhance the user experience by providing more options and flexibility. It would be particularly useful for users who need to evaluate different LLM providers for specific tasks or who want to ensure they are getting the best possible performance and accuracy. Additionally, supporting multiple providers could attract a wider range of users who have different preferences or requirements for their LLM interactions.
Author
Owner

@harryzhou2000 commented on GitHub (Feb 21, 2025):

You should be able to add multiple OpenAI APIs in the Connections tab of Admin Settings.

You can also add name prefixes to distinguish the models imported and restrict the set of models imported for each OpenAI API entry.

Does this solve the problem?


The settings page of version 0.5.16:

Image

<!-- gh-comment-id:2674449302 --> @harryzhou2000 commented on GitHub (Feb 21, 2025): You should be able to add multiple OpenAI APIs in the `Connections` tab of `Admin Settings`. You can also add name prefixes to distinguish the models imported and restrict the set of models imported for each OpenAI API entry. Does this solve the problem? --- The settings page of version `0.5.16`: ![Image](https://github.com/user-attachments/assets/5a2ecdb1-2040-463b-8f99-f1a71bc3c600)
Author
Owner

@tjbck commented on GitHub (Feb 21, 2025):

Functions.

<!-- gh-comment-id:2675202519 --> @tjbck commented on GitHub (Feb 21, 2025): Functions.
Author
Owner

@rzafiamy commented on GitHub (Feb 21, 2025):

@harryzhou2000 Indeed , it' s already available !
Very interresting feature ! Thanks for your quick reply.

Screenshot_20250221_194410_Firefox.jpg

<!-- gh-comment-id:2675296428 --> @rzafiamy commented on GitHub (Feb 21, 2025): @harryzhou2000 Indeed , it' s already available ! Very interresting feature ! Thanks for your quick reply. ![Screenshot_20250221_194410_Firefox.jpg](https://github.com/user-attachments/assets/76fde68a-1086-4dcf-b456-94ea791774cf)
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#86648