mirror of
https://github.com/open-webui/open-webui.git
synced 2026-07-13 19:43:35 -05:00
[GH-ISSUE #1486] feat: detailed model info #12518
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#12518
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 @sammcj on GitHub (Apr 10, 2024).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/1486
Bug Report
Description
Bug Summary:
The UI doesn't show which server is providing a given model.
For example, say you have:
And you're selecting which model to use - you have no idea which host you're going to be performing the inference on.
For example, in the following screenshot these models could be provided by OpenAI, OpenRouter, Ollama or some other proxy - and if there's multiple accounts/keys for a service - which one is in use?
Steps to Reproduce:
Add multiple providers with the same or similar model names and try to select a model to use.
Another nice option might be to give servers a 'name' attribute and show that.
Expected Behaviour:
In the models drop down, I'd expect next to the model name to have the server shown.
For example:
Actual Behaviour:
No model server is shown.
Environment
Reproduction Details
Confirmation:
Logs and Screenshots
Screenshots (if applicable):
Installation Method
n/a
Additional Information
n/a
Note
n/a
@justinh-rahb commented on GitHub (Apr 10, 2024):
Ya, tbh I'd kinda like if the "Assistant" profile picture in chats were instead the logo we set for a given model (defaults can be inferred by API base URLs and/or model names in many cases) too, along with other things that have been requested by others such as being able to edit names, set rate limits (per user), and other more granular controls for model and API access.
@sammcj commented on GitHub (Apr 10, 2024):
Using the Assistant profile picture would tie in nicely if Assistants end up supporting multiple models/servers as well ;)
@abqareno commented on GitHub (Apr 13, 2024):
i noticed also that [system prompt] is not based on which model in use, instead of is just one per chat. so we can't switch models in between.
@sammcj commented on GitHub (Apr 13, 2024):
@abqareno yeah I noticed that as well, I think that's worth logging a separate bug / feature request for though if you could?
@tjbck commented on GitHub (Apr 14, 2024):
@abqareno blocker: https://github.com/open-webui/open-webui/issues/665
@tjbck commented on GitHub (May 26, 2024):
Closing in favour of #1655