Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 14:13:35 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #3567] The dialogue effect in the workspace through the front-end UI is inconsistent with the API access effect, and the gap is quite large #2299

New Issue
Closed
opened 2026-02-22 18:29:04 -05:00 by yindo · 1 comment
yindo commented 2026-02-22 18:29:04 -05:00
Owner
Copy Link
Copy Source

Originally created by @Donaghys on GitHub (Mar 31, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3567

Thank you for the author's analysis work. I have been studying this project recently and have found that using the same model to create a workspace, and then uploading several documents as vertical knowledge of the model, to address the same issue, the feedback effect of the model is relatively good when conversing with the workspace through the front-end UI. However, when conversing with the model through API calls, the model effect is relatively poor. What is the problem?

Originally created by @Donaghys on GitHub (Mar 31, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3567 Thank you for the author's analysis work. I have been studying this project recently and have found that using the same model to create a workspace, and then uploading several documents as vertical knowledge of the model, to address the same issue, the feedback effect of the model is relatively good when conversing with the workspace through the front-end UI. However, when conversing with the model through API calls, the model effect is relatively poor. What is the problem?
yindo closed this issue 2026-02-22 18:29:04 -05:00
yindo commented 2026-02-22 18:29:04 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Mar 31, 2025):

Are you saying the exact same query from the API to a workspace and via the frontend generate very different and incorrect responses? They use the same process and document set for search so their responses may change slightly since LLMs are non-deterministic, but it should not be far apart in accuracy.

Can you provide some examples here that would be able to be reproduced? The flows, as far as code goes, are identical.

@timothycarambat commented on GitHub (Mar 31, 2025): Are you saying the exact same query from the API to a workspace and via the frontend generate _very different and incorrect_ responses? They use the same process and document set for search so their responses may change slightly since LLMs are non-deterministic, but it should not be far apart in accuracy. Can you provide some examples here that would be able to be reproduced? The flows, as far as code goes, are identical.
yindo changed title from The dialogue effect in the workspace through the front-end UI is inconsistent with the API access effect, and the gap is quite large to [GH-ISSUE #3567] The dialogue effect in the workspace through the front-end UI is inconsistent with the API access effect, and the gap is quite large 2026-06-05 14:45:45 -04:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
Desktop
Docker
Integration Request
Integration Request
OS: Linux
OS: Mobile
OS: Windows
UI/UX
blocked
bug
bug
core-team-only
documentation
duplicate
embed-widget
enhancement
feature request
github_actions
good first issue
investigating
needs info / can't replicate
possible bug
pull-request
Mirrored from GitHub Pull Request
 question
stage: specifications
wontfix
No Label
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: Mintplex-Labs/anything-llm#2299
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?