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 #3466] [BUG]: Workspace Chat mode Chat/Query Switch doesn't seems to be working #2236

New Issue
Closed
opened 2026-02-22 18:28:47 -05:00 by yindo · 2 comments
yindo commented 2026-02-22 18:28:47 -05:00
Owner
Copy Link
Copy Source

Originally created by @huangzijian037 on GitHub (Mar 14, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3466

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Anything LLM Desktop App v1.7.5

Workspace configuration:

LLM Provider: OpenAI API, model GPT-4
Chat Mode: Query
LLM Temperature: 0.3

All other settings deafult

Workspace document uploaded: Physical Activity Guidelines for Americans 2nd edition

In default thread, I ask two questions:

  1. How many physical activity should I get
    Expected answer: summarized response from the document I uploaded
    AnythingLLM Output: relavent response using citation from my document (not very ideal though)

  2. Why is sky blue
    Expected answer: There is no relevant information in this workspace to answer your query. (Deafult setting)
    AnythingLLM Output: The sky appears blue due to a phenomenon called Rayleigh scattering. When sunlight enters Earth’s atmosphere, it collides with molecules and small particles in the air. Sunlight is made up of different colors, each with different wavelengths. Blue light has a shorter wavelength and is scattered in all directions by the gases and particles in the atmosphere more than other colors. This scattered blue light is what we see when we look up at the sky.

Are there known steps to reproduce?

Please refer to the last section for how to reproduce

Originally created by @huangzijian037 on GitHub (Mar 14, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/3466 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Anything LLM Desktop App v1.7.5 Workspace configuration: LLM Provider: OpenAI API, model GPT-4 Chat Mode: Query LLM Temperature: 0.3 All other settings deafult _________________________________________________ Workspace document uploaded: Physical Activity Guidelines for Americans 2nd edition In default thread, I ask two questions: 1. How many physical activity should I get Expected answer: summarized response from the document I uploaded AnythingLLM Output: relavent response using citation from my document (not very ideal though) 3. Why is sky blue Expected answer: There is no relevant information in this workspace to answer your query. (Deafult setting) AnythingLLM Output: The sky appears blue due to a phenomenon called Rayleigh scattering. When sunlight enters Earth’s atmosphere, it collides with molecules and small particles in the air. Sunlight is made up of different colors, each with different wavelengths. Blue light has a shorter wavelength and is scattered in all directions by the gases and particles in the atmosphere more than other colors. This scattered blue light is what we see when we look up at the sky. ### Are there known steps to reproduce? Please refer to the last section for how to reproduce
yindo added the possible bug label 2026-02-22 18:28:47 -05:00
yindo closed this issue 2026-02-22 18:28:47 -05:00
yindo commented 2026-02-22 18:28:47 -05:00
Author
Owner
Copy Link
Copy Source

@shatfield4 commented on GitHub (Mar 14, 2025):

This is just due to how RAG works from a technical standpoint. The only way a chat will respond if in query mode will be if chunks of the text from a document are found. The questions you are asking to test this are very vague and will often times find chunks of text related to that very easily which is why it keeps responding even in query mode. You can try increasing the Document similarity threshold under vector database settings in your workspace and this will filter out results better to match more closely to the question you are asking.

@shatfield4 commented on GitHub (Mar 14, 2025): This is just due to how RAG works from a technical standpoint. The only way a chat will respond if in query mode will be if chunks of the text from a document are found. The questions you are asking to test this are very vague and will often times find chunks of text related to that very easily which is why it keeps responding even in query mode. You can try increasing the `Document similarity threshold` under vector database settings in your workspace and this will filter out results better to match more closely to the question you are asking.
yindo commented 2026-02-22 18:28:47 -05:00
Author
Owner
Copy Link
Copy Source

@huangzijian037 commented on GitHub (Mar 15, 2025):

Thanks for responding. I did try to increase the Document similarity threshold. However, in the query mode, the some generated answers are either based on 0% match chuncks or no citation at all. I would expect it give me refusal message "There is no relevant information in this workspace to answer your query." (Deafult setting)

@huangzijian037 commented on GitHub (Mar 15, 2025): Thanks for responding. I did try to increase the Document similarity threshold. However, in the query mode, the some generated answers are either based on 0% match chuncks or no citation at all. I would expect it give me refusal message "There is no relevant information in this workspace to answer your query." (Deafult setting)
yindo changed title from [BUG]: Workspace Chat mode Chat/Query Switch doesn't seems to be working to [GH-ISSUE #3466] [BUG]: Workspace Chat mode Chat/Query Switch doesn't seems to be working 2026-06-05 14:45:22 -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#2236
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?