[GH-ISSUE #23544] issue: Wrong Model selected! #20010

Closed
opened 2026-04-20 02:35:35 -05:00 by GiteaMirror · 2 comments
Owner

Originally created by @JLim-IRM on GitHub (Apr 9, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/23544

Check Existing Issues

  • I have searched for any existing and/or related issues.
  • I have searched for any existing and/or related discussions.
  • I have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!).
  • I am using the latest version of Open WebUI.

Installation Method

Git Clone

Open WebUI Version

v0.8.10

Ollama Version (if applicable)

"org.opencontainers.image.ref.name": "ubuntu", "org.opencontainers.image.version": "24.04"

Operating System

Windows 11 Enterprise

Browser (if applicable)

Edge, Version 146.0.3856.72 (Official build) (64-bit)

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

Prompt should ask the agent that was selected instead of mismatch due to bug.

Actual Behavior

It was labeled as one agent (for example) Claude, and I've asked who are you to identify it and it prompt it back as Gemini because it was selected in the last model for adding multiple models and adding @ in the chat to ask different agent.

Steps to Reproduce

Add a model and selected GPT.
In chat, add @ and select different agent like Gemini.
Add another model like different version of GPT.
In chat add @ and select another agent like Claude.
Add another model and in chat, @ to select something else.

Ask in prompt who are you and the label shows current or last selected agent, but it calls different agent.

Logs & Screenshots

Image Image

Additional Information

No response

Originally created by @JLim-IRM on GitHub (Apr 9, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/23544 ### 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 have also searched in the CLOSED issues AND CLOSED discussions and found no related items (your issue might already be addressed on the development branch!). - [x] I am using the latest version of Open WebUI. ### Installation Method Git Clone ### Open WebUI Version v0.8.10 ### Ollama Version (if applicable) "org.opencontainers.image.ref.name": "ubuntu", "org.opencontainers.image.version": "24.04" ### Operating System Windows 11 Enterprise ### Browser (if applicable) Edge, Version 146.0.3856.72 (Official build) (64-bit) ### 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 Prompt should ask the agent that was selected instead of mismatch due to bug. ### Actual Behavior It was labeled as one agent (for example) Claude, and I've asked who are you to identify it and it prompt it back as Gemini because it was selected in the last model for adding multiple models and adding @ in the chat to ask different agent. ### Steps to Reproduce Add a model and selected GPT. In chat, add @ and select different agent like Gemini. Add another model like different version of GPT. In chat add @ and select another agent like Claude. Add another model and in chat, @ to select something else. Ask in prompt who are you and the label shows current or last selected agent, but it calls different agent. ### Logs & Screenshots <img width="614" height="299" alt="Image" src="https://github.com/user-attachments/assets/d6c3fbce-1e25-47c7-ab21-d3699b3b2e38" /> <img width="569" height="242" alt="Image" src="https://github.com/user-attachments/assets/a8c82b71-bf24-4a37-8744-834857ca586c" /> ### Additional Information _No response_
GiteaMirror added the bug label 2026-04-20 02:35:35 -05:00
Author
Owner

@Classic298 commented on GitHub (Apr 9, 2026):

I tried to reproduce using the steps to reproduce but i am confused on some of the steps. Could you describe them better? I cannot follow them

<!-- gh-comment-id:4214701072 --> @Classic298 commented on GitHub (Apr 9, 2026): I tried to reproduce using the steps to reproduce but i am confused on some of the steps. Could you describe them better? I cannot follow them
Author
Owner

@JLim-IRM commented on GitHub (Apr 9, 2026):

@Classic298
step 1

Select a model then type @ in chat to switch to another model inside of the chat.

Image

step 2
Add another model and type @ to switch the agent again.

Image

step 3
Add another model and type @ to select new agent again.

Image

step 4
ask the agent who are you to test and it will point to the one before this one.

Image
<!-- gh-comment-id:4214843923 --> @JLim-IRM commented on GitHub (Apr 9, 2026): @Classic298 step 1 Select a model then type @ in chat to switch to another model inside of the chat. <img width="647" height="385" alt="Image" src="https://github.com/user-attachments/assets/0251f796-b2b0-4d3f-837d-aef591a96ae4" /> step 2 Add another model and type @ to switch the agent again. <img width="1333" height="816" alt="Image" src="https://github.com/user-attachments/assets/e9325d02-00d0-4685-9740-0ae98e78a3a9" /> step 3 Add another model and type @ to select new agent again. <img width="643" height="436" alt="Image" src="https://github.com/user-attachments/assets/1f401f70-7df1-42dc-a1ab-0e2b1823289a" /> step 4 ask the agent who are you to test and it will point to the one before this one. <img width="670" height="253" alt="Image" src="https://github.com/user-attachments/assets/90ed2e91-6041-4c4f-a337-9cfbdf1e7b37" />
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/open-webui#20010