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 #5556] [BUG]: @agent queries don't go anywhere [LMStudio] #5162

New Issue
Closed
opened 2026-06-05 14:52:23 -04:00 by yindo · 1 comment
yindo commented 2026-06-05 14:52:23 -04:00
Owner
Copy Link
Copy Source

Originally created by @Anonymous-1901 on GitHub (Apr 29, 2026).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5556

How are you running AnythingLLM?

AnythingLLM desktop app

What happened?

Hi,
I recently learned about Anything LLM, and wanted to give it a try. When I tried to connect to my existing LM Studio Models, it just didn't work at first.
But then when I toggled the Auto Mode to either Chat or Query in Workspace settings, I finally got a response. But when I use @agent it goes into the same void loop.

Image

See below, no requests being made to LM Studio:

Image

So basically I can just chat with models fine, but not use agentic capablities somehow.

Please let me know if I'm missing something since I'm new to the community.

Thanks.

Are there known steps to reproduce?

Connect to a model via LM Studio on the AnythingLLM Windows x64 Desktop Application, and run @agent commands, or Auto mode in Workspace Settings.

LLM Provider & Model (if applicable)

LM Studio / qwen3.5-2b , gpt-oss-20b, qwen3.6-35b-a3b

Embedder Provider & Model (if applicable)

LM Studio / text-embedding-nomic-embed-text-v1.5

Originally created by @Anonymous-1901 on GitHub (Apr 29, 2026). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/5556 ### How are you running AnythingLLM? AnythingLLM desktop app ### What happened? Hi, I recently learned about Anything LLM, and wanted to give it a try. When I tried to connect to my existing LM Studio Models, it just didn't work at first. But then when I toggled the Auto Mode to either Chat or Query in Workspace settings, I finally got a response. But when I use @agent it goes into the same void loop. <img width="2372" height="828" alt="Image" src="https://github.com/user-attachments/assets/614bbacb-f2a7-45c2-881e-5efc97305905" /> See below, no requests being made to LM Studio: <img width="2555" height="1460" alt="Image" src="https://github.com/user-attachments/assets/ce05b012-8e9b-4324-a8e9-9b6676eaa8d7" /> So basically I can just chat with models fine, but not use agentic capablities somehow. Please let me know if I'm missing something since I'm new to the community. Thanks. ### Are there known steps to reproduce? Connect to a model via LM Studio on the AnythingLLM Windows x64 Desktop Application, and run @agent commands, or Auto mode in Workspace Settings. ### LLM Provider & Model (if applicable) LM Studio / qwen3.5-2b , gpt-oss-20b, qwen3.6-35b-a3b ### Embedder Provider & Model (if applicable) LM Studio / text-embedding-nomic-embed-text-v1.5
yindo added the possible bug label 2026-06-05 14:52:23 -04:00
yindo closed this issue 2026-06-05 14:52:23 -04:00
yindo commented 2026-06-05 14:52:24 -04:00
Author
Owner
Copy Link
Copy Source

@timothycarambat commented on GitHub (Apr 29, 2026):

But then when I toggled the Auto Mode to either Chat or Query in Workspace settings, I finally got a response. But when I use https://github.com/agent it goes into the same void loop.

Do you happen to have a lot of agent skills enabled? Something that can happen with local models in general is adding tons of tools to their chat and the prefill take forever (and even possibly crash!) which is why going to chat may have sent a request and that worked.

Seems like you might have a lot of "agent skills" enabled. One tip to get around this is to enable Intelligent Tool Calling which is very new and why it is off by default. However, it will save you a TON of time on inference.

Let me know if that does/doesnt work

<!-- gh-comment-id:4348191544 --> @timothycarambat commented on GitHub (Apr 29, 2026): > But then when I toggled the Auto Mode to either Chat or Query in Workspace settings, I finally got a response. But when I use https://github.com/agent it goes into the same void loop. Do you happen to have a lot of agent skills enabled? Something that can happen with local models in general is adding tons of tools to their chat and the prefill take forever (and even possibly crash!) which is why going to chat may have sent a request and that worked. Seems like you might have a lot of "agent skills" enabled. One tip to get around this is to enable [Intelligent Tool Calling](https://docs.anythingllm.com/agent/intelligent-tool-selection#how-to-enable-intelligent-tool-selection) which is very new and why it is off by default. However, it will save you a TON of time on inference. Let me know if that does/doesnt work
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#5162
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?