[GH-ISSUE #23125] issue: open-terminal read_file inconsistent behavior between user-mode and server-mode #90628

New Issue

2026-05-15T15:54:11-05:00

GiteaMirror commented

2026-05-15 15:54:11 -05:00

Originally created by @Arokha on GitHub (Mar 27, 2026).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/23125

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

Docker

Open WebUI Version

dev

Ollama Version (if applicable)

No response

Operating System

ubuntu 22.04

Browser (if applicable)

Firefox

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

read_file follows through a path in backend/utils/tools.py that will allow an LLM to read images with it, by examining the Content-Type header. But, this only applies to open-terminal connections set up on a server-side basis.

Actual Behavior

Open-terminal connected on a user-side basis (in user settings) run res.text() in the browser on all responses from the open-terminal API, and lose the Content-Type header, and can never reach the path that would re-parse the binary into base64 for an LLM to use computer vision on. So, this feature seems only available for 'half' of the uses of open-terminal.

Steps to Reproduce

Configure open-terminal on server-side vs user-side and compare the results of read_file through various code paths.

Logs & Screenshots

See above

Additional Information

No response

Originally created by @Arokha on GitHub (Mar 27, 2026). Original GitHub issue: https://github.com/open-webui/open-webui/issues/23125 ### 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 Docker ### Open WebUI Version dev ### Ollama Version (if applicable) _No response_ ### Operating System ubuntu 22.04 ### Browser (if applicable) Firefox ### 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 read_file follows through a path in backend/utils/tools.py that will allow an LLM to read images with it, by examining the Content-Type header. But, this only applies to open-terminal connections set up on a server-side basis. ### Actual Behavior Open-terminal connected on a user-side basis (in user settings) run res.text() in the browser on all responses from the open-terminal API, and lose the Content-Type header, and can never reach the path that would re-parse the binary into base64 for an LLM to use computer vision on. So, this feature seems only available for 'half' of the uses of open-terminal. ### Steps to Reproduce Configure open-terminal on server-side vs user-side and compare the results of read_file through various code paths. ### Logs & Screenshots See above ### Additional Information _No response_

GiteaMirror added the bug label 2026-05-15 15:54:11 -05:00

GiteaMirror closed this issue

2026-05-15 15:54:14 -05:00

GiteaMirror commented

2026-05-15 15:54:15 -05:00

@tjbck commented on GitHub (Apr 1, 2026):

Addressed in dev.

@tjbck commented on GitHub (Apr 1, 2026): Addressed in dev.

Sign in to join this conversation.

Branches Tags

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/open-webui#90628