Mintplex-Labs/anything-llm
1
0
Fork 0
mirror of https://github.com/Mintplex-Labs/anything-llm.git synced 2026-07-19 22:23:50 -04:00
Code Issues 328 Packages Projects Releases 10 Wiki Activity

[GH-ISSUE #4798] [BUG]: Agent uses wrong model; fails with 404 #3020

New Issue
Closed
opened 2026-02-22 18:32:17 -05:00 by yindo · 2 comments
yindo commented 2026-02-22 18:32:17 -05:00
Owner
Copy Link
Copy Source

Originally created by @socketbox on GitHub (Dec 19, 2025).
Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4798

How are you running AnythingLLM?

Docker (local)

What happened?

I select a model (Anthropic Sonnet 4.5 for example, but this occurs no matter what I choose) using the "brain icon." I type a prompt using @agent. After hitting Enter, there's a 4 to 8 second pause, and then the interface shows this:

Image

This happened initially on 1.9.0. Then I updated to 1.9.1 and it's still occurring.

Are there known steps to reproduce?

On my machine, it happens every time I invoke the agent. In other environments...?

Stack trace:

[backend] info: [AgentHandler] [DEBUG] Provider supports agent streaming - will use async execution!
[backend] error:    error: 404 {"type":"error","error":{"type":"not_found_error","message":"model: claude-3-5-sonnet-latest"},"request_id":"req_011CWGtxxvBSCtEp3jFfEo7t"} Error: 404 {"type":"error","error":{"type":"not_found_error","message":"model: claude-3-5-sonnet-latest"},"request_id":"req_011CWGtxxvBSCtEp3jFfEo7t"}
    at AnthropicProvider.stream (/app/server/utils/agents/aibitat/providers/anthropic.js:299:15)
    at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
    at async AIbitat.handleAsyncExecution (/app/server/utils/agents/aibitat/index.js:651:30)
    at async AIbitat.reply (/app/server/utils/agents/aibitat/index.js:607:17)
    at async AIbitat.chat (/app/server/utils/agents/aibitat/index.js:388:15)
    at async AIbitat.start (/app/server/utils/agents/aibitat/index.js:322:5)
    at async /app/server/endpoints/agentWebsocket.js:52:7
[backend] info: [AgentHandler] End e2505a74-01eb-4833-9d14-5606f0cb2e9e::anthropic:claude-3-5-sonnet-latest
Originally created by @socketbox on GitHub (Dec 19, 2025). Original GitHub issue: https://github.com/Mintplex-Labs/anything-llm/issues/4798 ### How are you running AnythingLLM? Docker (local) ### What happened? I select a model (Anthropic Sonnet 4.5 for example, but this occurs no matter what I choose) using the "brain icon." I type a prompt using `@agent`. After hitting Enter, there's a 4 to 8 second pause, and then the interface shows this: <img width="1190" height="257" alt="Image" src="https://github.com/user-attachments/assets/6b9ad540-ea8b-4404-97d5-0ed7b32125f9" /> This happened initially on 1.9.0. Then I updated to 1.9.1 and it's still occurring. ### Are there known steps to reproduce? On my machine, it happens every time I invoke the agent. In other environments...? ### Stack trace: ``` [backend] info: [AgentHandler] [DEBUG] Provider supports agent streaming - will use async execution! [backend] error: error: 404 {"type":"error","error":{"type":"not_found_error","message":"model: claude-3-5-sonnet-latest"},"request_id":"req_011CWGtxxvBSCtEp3jFfEo7t"} Error: 404 {"type":"error","error":{"type":"not_found_error","message":"model: claude-3-5-sonnet-latest"},"request_id":"req_011CWGtxxvBSCtEp3jFfEo7t"} at AnthropicProvider.stream (/app/server/utils/agents/aibitat/providers/anthropic.js:299:15) at process.processTicksAndRejections (node:internal/process/task_queues:95:5) at async AIbitat.handleAsyncExecution (/app/server/utils/agents/aibitat/index.js:651:30) at async AIbitat.reply (/app/server/utils/agents/aibitat/index.js:607:17) at async AIbitat.chat (/app/server/utils/agents/aibitat/index.js:388:15) at async AIbitat.start (/app/server/utils/agents/aibitat/index.js:322:5) at async /app/server/endpoints/agentWebsocket.js:52:7 [backend] info: [AgentHandler] End e2505a74-01eb-4833-9d14-5606f0cb2e9e::anthropic:claude-3-5-sonnet-latest ```
yindo added the possible bug label 2026-02-22 18:32:17 -05:00
yindo closed this issue 2026-02-22 18:32:17 -05:00
yindo commented 2026-02-22 18:32:17 -05:00
Author
Owner
Copy Link
Copy Source

@socketbox commented on GitHub (Dec 19, 2025):

Looking at the stack trace, and engaging in pure conjecture, it might be that there is a stale request in the task queue that was initially made with Sonnet 3.5 and then abruptly interrupted. Now that 3.5 is no longer available, it can't be completed.

@socketbox commented on GitHub (Dec 19, 2025): Looking at the stack trace, and engaging in pure conjecture, it might be that there is a stale request in the task queue that was initially made with Sonnet 3.5 and then abruptly interrupted. Now that 3.5 is no longer available, it can't be completed.
yindo commented 2026-02-22 18:32:17 -05:00
Author
Owner
Copy Link
Copy Source

@shatfield4 commented on GitHub (Dec 22, 2025):

Image

Under the workspace settings here and on the agent configuration tab do you have a model set here? It may be trying to get the model from here and overriding the one you have set in your LLM configuration.

Note: Using the brain icon pop up window for setting the LLM only changes the chat LLM for that workspace. You can confirm this by going to workspace settings and the chat settings tab. It will update this section when changing via the brain icon popup. This does not change the model for agents. You have to go into the agent configuration tab to set it manually, or go into the main LLM settings (wrench icon in side bar) to switch every single workspace over to the selected model.

@shatfield4 commented on GitHub (Dec 22, 2025): <img width="1514" height="882" alt="Image" src="https://github.com/user-attachments/assets/b2d1c0cb-dbcc-48e6-b7c9-4c302a9c328e" /> Under the workspace settings here and on the agent configuration tab do you have a model set here? It may be trying to get the model from here and overriding the one you have set in your LLM configuration. Note: Using the brain icon pop up window for setting the LLM **only changes the chat LLM for that workspace**. You can confirm this by going to workspace settings and the chat settings tab. It will update this section when changing via the brain icon popup. This does not change the model for agents. You have to go into the agent configuration tab to set it manually, or go into the main LLM settings (wrench icon in side bar) to switch every single workspace over to the selected model.
yindo changed title from [BUG]: Agent uses wrong model; fails with 404 to [GH-ISSUE #4798] [BUG]: Agent uses wrong model; fails with 404 2026-06-05 14:49:51 -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#3020
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?