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 #4167] [BUG]: Reasoning (thinking) models response parsing #2651

New Issue
Closed
opened 2026-02-22 18:30:36 -05:00 by yindo · 2 comments
yindo commented 2026-02-22 18:30:36 -05:00
Owner
Copy Link
Copy Source

Originally created by @Fl0p on GitHub (Jul 17, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4167

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Trying to use several reasoning models with AnythingLLM desktop App .
No one succeeded

Grok4 (with openRouter) almost each time stuck on <Thinking...>
Image

Gemmini2.5Pro (with openRouter) almost each time stuck in the middle of reasoning
Image

OpenAI o1Pro (with openRouter) just stuck each time
Image

OpenAI o1Pro (Direct) not supported API
Image

Are there known steps to reproduce?

  • Just try to use any reasoning(thinking) models.
  • observe the result
Originally created by @Fl0p on GitHub (Jul 17, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4167 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Trying to use several reasoning models with AnythingLLM desktop App . No one succeeded Grok4 (with openRouter) almost each time stuck on <Thinking...> <img width="989" height="804" alt="Image" src="https://github.com/user-attachments/assets/b50efd2c-5c01-41d6-84f7-4b673bf195fd" /> Gemmini2.5Pro (with openRouter) almost each time stuck in the middle of reasoning <img width="1079" height="853" alt="Image" src="https://github.com/user-attachments/assets/145db2f1-b58f-4030-8fe4-6563b37dd1a9" /> OpenAI o1Pro (with openRouter) just stuck each time <img width="626" height="277" alt="Image" src="https://github.com/user-attachments/assets/7720c970-e713-459c-b41f-48e28e112344" /> OpenAI o1Pro (Direct) not supported API <img width="768" height="285" alt="Image" src="https://github.com/user-attachments/assets/02f16dbf-56cf-4f4f-aaa2-abafa72bcb83" /> ### Are there known steps to reproduce? - Just try to use any reasoning(thinking) models. - observe the result
yindo added the possible bug label 2026-02-22 18:30:36 -05:00
yindo closed this issue 2026-02-22 18:30:37 -05:00
yindo commented 2026-02-22 18:30:37 -05:00
Author
Owner
Copy Link
Copy Source

@Fl0p commented on GitHub (Jul 17, 2025):

BTW perplexity works fine

Image
@Fl0p commented on GitHub (Jul 17, 2025): BTW perplexity works fine <img width="1141" height="1014" alt="Image" src="https://github.com/user-attachments/assets/db23fa58-631d-453a-a252-004379c1d9e7" />
yindo commented 2026-02-22 18:30:37 -05:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Jul 17, 2025):

This is because we have a forced timeout limit for OpenRouter because they host some models that never return a close response. In the LLM Preferences for OpenRouter, expand the advanced options part and change it to something high like 5000.

What is happeneing is we are closing the response early since it goes more than 500ms between tokens so we assume the stream closed since OR wont tell us if it did or not. So in this case, you never get the ending <think> tag.

This stream/timeout thing does not apply to every provider/model, but its not clear why some do and some dont return a close message. If we dont force it closed you cannot send another prompt since the stream is assumed to still be open. That is why that even exists

@timothycarambat commented on GitHub (Jul 17, 2025): This is because we have a forced timeout limit for OpenRouter because they host _some_ models that never return a close response. In the LLM Preferences for OpenRouter, expand the advanced options part and change it to something high like 5000. What is happeneing is we are closing the response early since it goes more than 500ms between tokens so we assume the stream closed since OR wont tell us if it did or not. So in this case, you never get the ending `<think>` tag. This stream/timeout thing does not apply to every provider/model, but its not clear why some do and some dont return a close message. If we dont force it closed you cannot send another prompt since the stream is assumed to still be open. That is why that even exists
yindo changed title from [BUG]: Reasoning (thinking) models response parsing to [GH-ISSUE #4167] [BUG]: Reasoning (thinking) models response parsing 2026-06-05 14:47:46 -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#2651
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?