Originally created by @Bickio on GitHub (Oct 20, 2025).
Check Existing Issues
I have searched for any existing and/or related issues.
I have searched for any existing and/or related discussions.
I am using the latest version of Open WebUI.
Installation Method
Docker
Open WebUI Version
v0.6.34
Ollama Version (if applicable)
No response
Operating System
MacOS 15.4.1
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
Model does its initial thinking, uses the tool natively and then continues its response
Actual Behavior
Model output stops after the first tool use
Steps to Reproduce
Configure a reasoning + native function execution model in litellm (I tested with Claude 4.5 Sonnet)
Use merge_reasoning_content_in_choices: true, but no other special settings
Configure litellm as an openai server in openwebui
Adjust the model settings in openwebui (either in admin, or chat settings):
Set Function Calling to Native
Set Reasoning Effort to medium
Configure an "external tool" connection in openwebui
Prompt the model with the tool enabled, in such a way that it will use the tool
Logs & Screenshots
SCREENSHOT FROM v0.6.34 (NOT WORKING):
SCREENSHOT FROM v0.6.33 (WORKING):
Additional Information
I've tested up and downgrading the versions multiple times, and rerunning the prompts multiple times. This is 100% reproducible and clearly a new regression between v0.6.33 and v0.6.34
Originally created by @Bickio on GitHub (Oct 20, 2025).
### Check Existing Issues
- [x] I have searched for any existing and/or related issues.
- [x] I have searched for any existing and/or related discussions.
- [x] I am using the latest version of Open WebUI.
### Installation Method
Docker
### Open WebUI Version
v0.6.34
### Ollama Version (if applicable)
_No response_
### Operating System
MacOS 15.4.1
### 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
Model does its initial thinking, uses the tool natively and then continues its response
### Actual Behavior
Model output stops after the first tool use
### Steps to Reproduce
- Configure a reasoning + native function execution model in litellm (I tested with Claude 4.5 Sonnet)
- Use `merge_reasoning_content_in_choices: true`, but no other special settings
- Configure litellm as an openai server in openwebui
- Adjust the model settings in openwebui (either in admin, or chat settings):
- Set `Function Calling` to `Native`
- Set `Reasoning Effort` to `medium`
- Configure an "external tool" connection in openwebui
- Prompt the model with the tool enabled, in such a way that it will use the tool
### Logs & Screenshots
SCREENSHOT FROM v0.6.34 (NOT WORKING):
<img width="711" height="153" alt="Image" src="https://github.com/user-attachments/assets/4b96887d-a061-46e1-b3e1-3fe944f0f060" />
SCREENSHOT FROM v0.6.33 (WORKING):
<img width="986" height="414" alt="Image" src="https://github.com/user-attachments/assets/a9d8be49-7cbe-42dc-a190-cdaf2bac6171" />
### Additional Information
I've tested up and downgrading the versions multiple times, and rerunning the prompts multiple times. This is 100% reproducible and clearly a new regression between v0.6.33 and v0.6.34
GiteaMirror
added the bug label 2025-11-11 17:04:00 -06:00
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @Bickio on GitHub (Oct 20, 2025).
Check Existing Issues
Installation Method
Docker
Open WebUI Version
v0.6.34
Ollama Version (if applicable)
No response
Operating System
MacOS 15.4.1
Browser (if applicable)
No response
Confirmation
README.md.Expected Behavior
Model does its initial thinking, uses the tool natively and then continues its response
Actual Behavior
Model output stops after the first tool use
Steps to Reproduce
Configure a reasoning + native function execution model in litellm (I tested with Claude 4.5 Sonnet)
merge_reasoning_content_in_choices: true, but no other special settingsConfigure litellm as an openai server in openwebui
Adjust the model settings in openwebui (either in admin, or chat settings):
Function CallingtoNativeReasoning EfforttomediumConfigure an "external tool" connection in openwebui
Prompt the model with the tool enabled, in such a way that it will use the tool
Logs & Screenshots
SCREENSHOT FROM v0.6.34 (NOT WORKING):
SCREENSHOT FROM v0.6.33 (WORKING):
Additional Information
I've tested up and downgrading the versions multiple times, and rerunning the prompts multiple times. This is 100% reproducible and clearly a new regression between v0.6.33 and v0.6.34
@tjbck commented on GitHub (Oct 20, 2025):
Backend logs required here, most likely has to do with the Pipe Function implementation.