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 #755] [BUG]: Response different between embed chat and workspace chat #438

New Issue
Closed
opened 2026-02-22 18:19:29 -05:00 by yindo · 2 comments
yindo commented 2026-02-22 18:19:29 -05:00
Owner
Copy Link
Copy Source

Originally created by @Mirgiacomo on GitHub (Feb 19, 2024).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/755

How are you running AnythingLLM?

Local development

What happened?

I noticed this anomaly:

I created a workspace where I uploaded some material. As a model I use GPT-3.5-turbo.
If from the platform where I create the workspaces I chat with the chatbot, and ask it questions that have nothing to do with it such as "How much does the sun weigh?", it responds correctly saying "I do not know the answer to that question. If you have any other questions related to Test test, feel free to ask! And remember, for any medical concerns, you can contact Test Hospital support at 000000 or e-mail test@test.it. To book a medical visit, you can visit: https://prenota.test.it/"
Which is the right answer I expect because I don't want to answer questions out of context.

image

Going to embed this workspace in an iframe in a site, however, I found that when writing the same question, the answer it provides is different:

image Which is not the answer I expect.

The workspace is the same, the prompt is therefore the same.
I don't understand why it gives different answers.

It almost seems that in the embed part it does not take the prompt into consideration.

Are there known steps to reproduce?

No response

Originally created by @Mirgiacomo on GitHub (Feb 19, 2024). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/755 ### How are you running AnythingLLM? Local development ### What happened? I noticed this anomaly: I created a workspace where I uploaded some material. As a model I use GPT-3.5-turbo. If from the platform where I create the workspaces I chat with the chatbot, and ask it questions that have nothing to do with it such as "How much does the sun weigh?", it responds correctly saying "_**I do not know the answer to that question. If you have any other questions related to Test test, feel free to ask! And remember, for any medical concerns, you can contact Test Hospital support at 000000 or e-mail test@test.it. To book a medical visit, you can visit: https://prenota.test.it/**_" Which is the right answer I expect because I don't want to answer questions out of context. <img width="886" alt="image" src="https://github.com/Mintplex-Labs/anything-llm/assets/22151717/6e0cfe17-232f-48c1-bc31-9904deb01559"> Going to embed this workspace in an iframe in a site, however, I found that when writing the same question, the answer it provides is different: <img width="410" alt="image" src="https://github.com/Mintplex-Labs/anything-llm/assets/22151717/7d85421b-90bf-44f1-b87d-d06c5cc305a7"> Which is not the answer I expect. The workspace is the same, the prompt is therefore the same. I don't understand why it gives different answers. It almost seems that in the embed part it does not take the prompt into consideration. ### Are there known steps to reproduce? _No response_
yindo added the possible bug label 2026-02-22 18:19:29 -05:00
yindo closed this issue 2026-02-22 18:19:29 -05:00
yindo commented 2026-02-22 18:19:29 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Feb 19, 2024):

Did you define a data-prompt property in the config and/or allow override of the system prompt in the config?

@timothycarambat commented on GitHub (Feb 19, 2024): Did you define a `data-prompt` property in the config and/or allow override of the system prompt in the config?
yindo commented 2026-02-22 18:19:29 -05:00
Author
Owner
Copy Link
Copy Source

@Mirgiacomo commented on GitHub (Feb 19, 2024):

Yes, it was not flashed.

After I cleared the cache it was working properly.... idk

@Mirgiacomo commented on GitHub (Feb 19, 2024): Yes, it was not flashed. After I cleared the cache it was working properly.... idk
yindo changed title from [BUG]: Response different between embed chat and workspace chat to [GH-ISSUE #755] [BUG]: Response different between embed chat and workspace chat 2026-06-05 14:35:18 -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 possible bug
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#438
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?