github-starred/open-webui
2
0
Fork 0
mirror of https://github.com/open-webui/open-webui.git synced 2026-05-06 10:58:17 -05:00
Code Issues 1.3k Packages Projects Releases 126 Wiki Activity

issue: Continuous Repetition of Tool-Use Responses #6133

New Issue
Closed
opened 2025-11-11 16:45:46 -06:00 by GiteaMirror · 2 comments
GiteaMirror commented 2025-11-11 16:45:46 -06:00
Owner
Copy Link

Originally created by @ziozzang on GitHub (Aug 19, 2025).

Check Existing Issues

  • I have searched the existing issues and discussions.
  • I am using the latest version of Open WebUI.

Installation Method

Docker

Open WebUI Version

0.6.22

Ollama Version (if applicable)

No response

Operating System

linux

Browser (if applicable)

No response

Confirmation

  • I have read and followed all instructions in README.md.
  • I am using the latest version of both Open WebUI and Ollama.
  • I have included the browser console logs.
  • I have included the Docker container logs.
  • I have provided every relevant configuration, setting, and environment variable used in my setup.
  • I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
  • I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
  • Start with the initial platform/version/OS and dependencies used,
  • Specify exact install/launch/configure commands,
  • List URLs visited, user input (incl. example values/emails/passwords if needed),
  • Describe all options and toggles enabled or changed,
  • Include any files or environmental changes,
  • Identify the expected and actual result at each stage,
  • Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

When I use GLM-4.5 with vLLM for tool-use, an unusual phenomenon has been observed. A specific phrase related to tool-use is continuously being repeated during the process.

Upon investigation, I have found that this issue occurs under the following conditions:

  1. User query.
  2. LLM response. (I will use a tool)
  3. Reasoning ()
  4. Tool-use and response.

Expected//

  • step 2. LLM response before tool-use print only once.
  • and 3-4(tool-use..) , 3-4(another tool use...) ...

Actual Behavior

When steps 3 and 4 are repeated, the response from step 2 is continuously output in between each repetition. Since this content does not exist in the actual API-level LLM request, it appears to be an issue with OpenWebUI's output handling. The message should be displayed only once and should not be repeatedly output.

Steps to Reproduce

docker v 0.6.22.

This is same issue.
https://www.reddit.com/r/LocalLLaMA/comments/1mekoy8/agentic_email_workflow_inside_of_openwebui/

Unlike other models, it appears that due to the specific characteristics of the GLM-4.5 model, it generates its current status outside of the tag during tool-use. Even with this model, there are no issues with the tool's invocation or usage, other than the repetition of the output on the OpenWebUI display.

Logs & Screenshots

no log. but posting with screenshot.

https://www.reddit.com/r/LocalLLaMA/comments/1mekoy8/agentic_email_workflow_inside_of_openwebui/

Additional Information

No response

Originally created by @ziozzang on GitHub (Aug 19, 2025). ### Check Existing Issues - [x] I have searched the existing issues and discussions. - [x] I am using the latest version of Open WebUI. ### Installation Method Docker ### Open WebUI Version 0.6.22 ### Ollama Version (if applicable) _No response_ ### Operating System linux ### Browser (if applicable) _No response_ ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior When I use GLM-4.5 with vLLM for tool-use, an unusual phenomenon has been observed. A specific phrase related to tool-use is continuously being repeated during the process. Upon investigation, I have found that this issue occurs under the following conditions: 1. User query. 2. LLM response. (I will use a tool) 3. Reasoning (<think>) 4. Tool-use and response. Expected// - step 2. LLM response before tool-use print only once. - and 3-4(tool-use..) , 3-4(another tool use...) ... ### Actual Behavior When steps 3 and 4 are repeated, the response from step 2 is continuously output in between each repetition. Since this content does not exist in the actual API-level LLM request, it appears to be an issue with OpenWebUI's output handling. The message should be displayed only once and should not be repeatedly output. ### Steps to Reproduce docker v 0.6.22. This is same issue. https://www.reddit.com/r/LocalLLaMA/comments/1mekoy8/agentic_email_workflow_inside_of_openwebui/ Unlike other models, it appears that due to the specific characteristics of the GLM-4.5 model, it generates its current status outside of the <think> tag during tool-use. Even with this model, there are no issues with the tool's invocation or usage, other than the repetition of the output on the OpenWebUI display. ### Logs & Screenshots no log. but posting with screenshot. https://www.reddit.com/r/LocalLLaMA/comments/1mekoy8/agentic_email_workflow_inside_of_openwebui/ ### Additional Information _No response_
GiteaMirror added the bug label 2025-11-11 16:45:46 -06:00
GiteaMirror commented 2025-11-11 16:45:47 -06:00
Author
Owner
Copy Link

@ziozzang commented on GitHub (Aug 19, 2025):

LLM level temporary solution

  • add system prompt like this
When using a tool, do not generate any responses until the tool usage is complete. All responses must be generated only after the usage is finished
@ziozzang commented on GitHub (Aug 19, 2025): LLM level temporary solution - add system prompt like this ``` When using a tool, do not generate any responses until the tool usage is complete. All responses must be generated only after the usage is finished ```
GiteaMirror commented 2025-11-11 16:45:47 -06:00
Author
Owner
Copy Link

@tjbck commented on GitHub (Aug 19, 2025):

Most likely an issue with the model itself, we're unable to reproduce with openai models. Keep us updated.

@tjbck commented on GitHub (Aug 19, 2025): Most likely an issue with the model itself, we're unable to reproduce with openai models. Keep us updated.
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
confirmed
confirmed issue
core
documentation
enhancement
good first issue
help wanted
non-core
pull-request
Mirrored from GitHub Pull Request
 python
question
testing wanted
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#6133
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?