[GH-ISSUE #18726] issue: web search and image generation

GiteaMirror commented

2026-04-20 00:53:50 -05:00

Owner

Originally created by @allmazz on GitHub (Oct 29, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/18726

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

Debian 12

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

If a model supports native tool calling(and it's enabled), Web Search and Image Generation exported to it as external tools, that it can run dynamically, as described here. No small models and search/image generation run until the model decide, it works in Native Mode.

Actual Behavior

It still behaves as in default mode: always running a small model before the main request and does Web Search or Image Generation.

Steps to Reproduce

Install and run latest WebUI in any way
Configure Web Search and Image Generation
Configure OpenAI API
Add GPT-4o model, enable native tool calling by default, enable Web Search and Image Generation as default features. Do not add custom instructions or touch any other settings.
Start a chat and say Hi to the model

Logs & Screenshots

Additional Information

The problem is old and known, and there are very popular workarounds on the Community site:

But anyway, all of them do not proxy all features and lacking in support of OCR, Image size/quality configuration, etc.

Originally created by @allmazz on GitHub (Oct 29, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/18726 ### 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 Debian 12 ### 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 If a model supports native tool calling(and it's enabled), Web Search and Image Generation exported to it as external tools, that it can run dynamically, [as described here](https://docs.openwebui.com/features/plugin/tools/#-native-mode-function-calling-built-in). No small models and search/image generation run until the model decide, it works in Native Mode. ### Actual Behavior It still behaves as in default mode: always running a small model before the main request and does Web Search or Image Generation. ### Steps to Reproduce 1. Install and run latest WebUI in any way 2. Configure Web Search and Image Generation 3. Configure OpenAI API 4. Add GPT-4o model, enable native tool calling by default, enable Web Search and Image Generation as default features. Do not add custom instructions or touch any other settings. 5. Start a chat and say Hi to the model ### Logs & Screenshots <img width="1041" height="601" alt="Image" src="https://github.com/user-attachments/assets/586898b0-c07c-4f37-81e9-3f50dfda8006" /> ### Additional Information The problem is old and known, and there are very popular workarounds on the Community site: * https://openwebui.com/t/justinrahb/image_gen * https://openwebui.com/t/eidenz/image_gen_filter_v2 * https://openwebui.com/t/nokodo/auto_web_search * https://openwebui.com/t/victor1203/web_search_with_google * https://openwebui.com/t/mamei16/llm_web_search But anyway, all of them do not proxy all features and lacking in support of OCR, Image size/quality configuration, etc.