mirror of
https://github.com/open-webui/open-webui.git
synced 2026-05-06 10:58:17 -05:00
[PR #20687] [CLOSED] feat: Sticky random model selection per conversation (pick once, lock for all turns) #64584
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?
📋 Pull Request Information
Original PR: https://github.com/open-webui/open-webui/pull/20687
Author: @ghost
Created: 1/15/2026
Status: ❌ Closed
Base:
dev← Head:main📝 Commits (10+)
fe6783cMerge pull request #19030 from open-webui/devfc05e0aMerge pull request #19405 from open-webui/deve3faec6Merge pull request #19416 from open-webui/dev9899293Merge pull request #19448 from open-webui/dev140605eMerge pull request #19462 from open-webui/dev6f1486fMerge pull request #19466 from open-webui/devd95f533Merge pull request #19729 from open-webui/deva7271530.6.43 (#20093)6adde20Merge pull request #20394 from open-webui/devf9b0534Merge pull request #20522 from open-webui/dev📊 Changes
5 files changed (+88 additions, -33 deletions)
View changed files
📝
backend/open_webui/config.py(+7 -0)📝
backend/open_webui/main.py(+3 -0)📝
backend/open_webui/routers/evaluations.py(+9 -0)📝
backend/open_webui/utils/chat.py(+63 -33)📝
src/lib/components/admin/Settings/Evaluations.svelte(+6 -0)📄 Description
Changelog Entry
Description
Implemented feature request #20668 by ankush13r. Added toggleable, turn persistent models per Arena chat. At the beginning of an Arena chat, one model is chosen for the duration of that chat. This can be disabled in Admin Panel -> Settings -> Evaluations. This allows models to be evaluated in Arena across multiple turns.
Added
Changed
Additional Information
This PR is not written by any AI agent.
Code Review has been performed, and functionality has been recorded:
The functionality of this feature was tested by recording five prompts with per-chat caching enabled and disabled. As both of these tests passed, I can verify this feature as working
Before (Or with per-chat caching disabled):
After (with per-chat caching enabled):
Contributor License Agreement
By submitting this pull request, I confirm that I have read and fully agree to the Contributor License Agreement (CLA), and I am providing my contributions under its terms.
🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.