[GH-ISSUE #23496] issue: "Continue" (Generate More) ignores edited assistant message and repeats original text

Originally created by @leetraman822 on GitHub (Apr 8, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/23496 ### 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 v0.8.12 ### Ollama Version (if applicable) _No response_ ### Operating System Windows 10 ### 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 edit an assistant's message in the chat and click the "Continue" (Generate More) button, the model should use the newly edited text as the current assistant payload and seamlessly append the newly generated completion to it. For example, if the AI says: Hello, nice to meet you, how can I help? And I edit it to: Hello, we meet again, today I need help with Clicking "Continue" should result in: Hello, we meet again, today I need help with [new generated content]. ### Actual Behavior The frontend UI ignores the edits made to the assistant's message. When "Continue" is triggered, the underlying request seems to use the original, unedited message state. Furthermore, instead of replacing or correctly appending, it repeats the original text. Using the example above, after editing and clicking "Continue", the bubble turns into: Hello, nice to meet you, how can I help?Hello, nice to meet you, how can I help? The edits are completely lost, and the original message is duplicated. This occurs regardless of the model or API backend being used. ### Steps to Reproduce 1.Start a chat with any model. 2.Send a prompt (e.g., "Say hello"). 3.Wait for the assistant to generate a response (e.g., "Hello, nice to meet you, how can I help you today?"). 4.Hover over the assistant's message and click the Edit button. 5.Modify the text significantly (e.g., change it to "Hello, we meet again, today I need help with"). 6.Click the checkmark to Save the edit. 7.Click the Continue button at the bottom of the assistant's message bubble. 8.Observe the result: The message reverts to the original text and duplicates it, ignoring the saved edits completely. ### Logs & Screenshots <img width="2150" height="526" alt="Image" src="https://github.com/user-attachments/assets/f6240f04-0f23-4c01-b310-cfefede45db6" /> <img width="1301" height="490" alt="Image" src="https://github.com/user-attachments/assets/c5794a79-2710-43e1-921a-d5f89d8991b8" /> <img width="1334" height="473" alt="Image" src="https://github.com/user-attachments/assets/c5597bcc-d16a-4a53-a2de-e6bb513d279a" /> <img width="1270" height="370" alt="Image" src="https://github.com/user-attachments/assets/fb90d92f-71d1-4792-abf6-e7992ec972a5" /> <img width="1271" height="458" alt="Image" src="https://github.com/user-attachments/assets/9b9daebe-764a-42eb-9bae-71a354ef67bb" /> ### Additional Information _No response_

GiteaMirror commented

2026-05-21 02:36:33 -05:00

Owner

Originally created by @leetraman822 on GitHub (Apr 8, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/23496

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

v0.8.12

Ollama Version (if applicable)

No response

Operating System

Windows 10

Browser (if applicable)

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 edit an assistant's message in the chat and click the "Continue" (Generate More) button, the model should use the newly edited text as the current assistant payload and seamlessly append the newly generated completion to it.

For example, if the AI says: Hello, nice to meet you, how can I help?
And I edit it to: Hello, we meet again, today I need help with
Clicking "Continue" should result in: Hello, we meet again, today I need help with [new generated content].

Actual Behavior

The frontend UI ignores the edits made to the assistant's message. When "Continue" is triggered, the underlying request seems to use the original, unedited message state.

Furthermore, instead of replacing or correctly appending, it repeats the original text.
Using the example above, after editing and clicking "Continue", the bubble turns into:
Hello, nice to meet you, how can I help?Hello, nice to meet you, how can I help?
The edits are completely lost, and the original message is duplicated. This occurs regardless of the model or API backend being used.