I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.
Installation Method
Docker
Open WebUI Version
v0.6.12
Ollama Version (if applicable)
No response
Operating System
Debian Linux 12
Browser (if applicable)
Doesn't matter
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
Hello,
I've been using OpenWebUI for a long time as my main AI provider (docker-compose). I've set up a few API's like OpenAI API, Groq cloud and Google AI Studio.
Everything has been working perfectly until I updated to the latest open-webui docker container (v0.6.12). The webui should request a response from google api and show the user the response.
Actual Behavior
The newest version breaks the google API support, both the website and container logs show: "400: bad request" error. I have verified The issue is not with the api, as it works on my pc (Archlinux), phone (Android; Termux) and a Fedora LXC on my proxmox server. The OpenAI and Groq API's continue to work though.
To resolve the issue i changed the container from open-webui:main to open-webui:v0.6.11 in my docker-compose file. It is working without any issues.
Steps to Reproduce
Set up the newest version of OpenWebUI using docker-compose
Add a Google API key to the webui
Try to run any model (eg. gemini-2.0-flash)
Get the "400: Bad request" error.
Logs & Screenshots
Additional Information
I don't believe the browser logs and the configuration are required as the older version of open-webui works just fine with the same configs and browsers.
Originally created by @Epickitrolaz on GitHub (May 29, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/14490
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
- [x] I am using the latest version of Open WebUI.
### Installation Method
Docker
### Open WebUI Version
v0.6.12
### Ollama Version (if applicable)
_No response_
### Operating System
Debian Linux 12
### Browser (if applicable)
Doesn't matter
### 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
Hello,
I've been using OpenWebUI for a long time as my main AI provider (docker-compose). I've set up a few API's like OpenAI API, Groq cloud and Google AI Studio.
Everything has been working perfectly until I updated to the latest open-webui docker container (v0.6.12). The webui should request a response from google api and show the user the response.
### Actual Behavior
The newest version breaks the google API support, both the website and container logs show: "400: bad request" error. I have verified The issue is not with the api, as it works on my pc (Archlinux), phone (Android; Termux) and a Fedora LXC on my proxmox server. The OpenAI and Groq API's continue to work though.
To resolve the issue i changed the container from open-webui:main to open-webui:v0.6.11 in my docker-compose file. It is working without any issues.
### Steps to Reproduce
* Set up the newest version of OpenWebUI using docker-compose
* Add a Google API key to the webui
* Try to run any model (eg. gemini-2.0-flash)
* Get the "400: Bad request" error.
### Logs & Screenshots
### Additional Information
I don't believe the browser logs and the configuration are required as the older version of open-webui works just fine with the same configs and browsers.
GiteaMirror
added the bug label 2026-05-20 22:13:03 -05:00
This was not fixed for me in 0.6.13. Google Gemini API still returns a 400, no matter which model I select, but local models work fine. I am using the default models, no system prompts in the model settings, or user settings, and all default. I tried making a new Admin user, but that didn't help. Rolling back to ghcr.io/open-webui/open-webui:git-b8e1621 fixed it (0.6.11). Why would this still be happening? Im using docker on Ubuntu. Any hints appreciated.
<!-- gh-comment-id:2921333497 -->
@VideoFX commented on GitHub (May 30, 2025):
400: Open WebUI: Server Connection Error
This was not fixed for me in 0.6.13. Google Gemini API still returns a 400, no matter which model I select, but local models work fine. I am using the default models, no system prompts in the model settings, or user settings, and all default. I tried making a new Admin user, but that didn't help. Rolling back to ghcr.io/open-webui/open-webui:git-b8e1621 fixed it (0.6.11). Why would this still be happening? Im using docker on Ubuntu. Any hints appreciated.
After the update all the Gemini, openai, ollama, groq and gpt4free models are working fine for me.
<!-- gh-comment-id:2921354914 -->
@Epickitrolaz commented on GitHub (May 30, 2025):
After the update all the Gemini, openai, ollama, groq and gpt4free models are working fine for me.
for me, 0.6.13 did not solve anything. I have the exact same error. my settings was fine before updating up to 0.6.12. I tried to modify the endpoint and url, settings, etc. without success. I've read that people experimenting this error have actually badly set the UI, but without finding the right parameters... I would be very interested in getting the actual good way to set the google api.
<!-- gh-comment-id:2956487322 -->
@EntryGoon commented on GitHub (Jun 9, 2025):
for me, 0.6.13 did not solve anything. I have the exact same error. my settings was fine before updating up to 0.6.12. I tried to modify the endpoint and url, settings, etc. without success. I've read that people experimenting this error have actually badly set the UI, but without finding the right parameters... I would be very interested in getting the actual good way to set the google api.
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 @Epickitrolaz on GitHub (May 29, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/14490
Check Existing Issues
Installation Method
Docker
Open WebUI Version
v0.6.12
Ollama Version (if applicable)
No response
Operating System
Debian Linux 12
Browser (if applicable)
Doesn't matter
Confirmation
README.md.Expected Behavior
Hello,
I've been using OpenWebUI for a long time as my main AI provider (docker-compose). I've set up a few API's like OpenAI API, Groq cloud and Google AI Studio.
Everything has been working perfectly until I updated to the latest open-webui docker container (v0.6.12). The webui should request a response from google api and show the user the response.
Actual Behavior
The newest version breaks the google API support, both the website and container logs show: "400: bad request" error. I have verified The issue is not with the api, as it works on my pc (Archlinux), phone (Android; Termux) and a Fedora LXC on my proxmox server. The OpenAI and Groq API's continue to work though.
To resolve the issue i changed the container from open-webui:main to open-webui:v0.6.11 in my docker-compose file. It is working without any issues.
Steps to Reproduce
Logs & Screenshots
Additional Information
I don't believe the browser logs and the configuration are required as the older version of open-webui works just fine with the same configs and browsers.
@tjbck commented on GitHub (May 29, 2025):
#14479
@VideoFX commented on GitHub (May 30, 2025):
400: Open WebUI: Server Connection Error
This was not fixed for me in 0.6.13. Google Gemini API still returns a 400, no matter which model I select, but local models work fine. I am using the default models, no system prompts in the model settings, or user settings, and all default. I tried making a new Admin user, but that didn't help. Rolling back to ghcr.io/open-webui/open-webui:git-b8e1621 fixed it (0.6.11). Why would this still be happening? Im using docker on Ubuntu. Any hints appreciated.
@Epickitrolaz commented on GitHub (May 30, 2025):
After the update all the Gemini, openai, ollama, groq and gpt4free models are working fine for me.
@EntryGoon commented on GitHub (Jun 9, 2025):
for me, 0.6.13 did not solve anything. I have the exact same error. my settings was fine before updating up to 0.6.12. I tried to modify the endpoint and url, settings, etc. without success. I've read that people experimenting this error have actually badly set the UI, but without finding the right parameters... I would be very interested in getting the actual good way to set the google api.