Originally created by @holzerseb on GitHub (Aug 4, 2025).
Check Existing Issues
I have searched the existing issues and discussions.
Problem Description
I am using Azure OpenAI for all models. With the GPT-image-1 and Dall-e-3 available on Azure OpenAI, we want to support image generation through azure. However, any attempts to configure the image generation with azure fails.
I have attempted to use the Azure OpenAI API URL instead of the default OpenAI URL and tested multiple options for the URL:
I wanted to test it with Dall-e-3 first, so my default model is currently configured with "dall-e-3". Image size is 1024x1024. Steps is set to 1, but I've tried 10, 50, 100 too.
I'm using 4o for the chat.
While Open WebUI saves and enables the feature correctly, I get this response:
Clicking on the "Generate image" button, results in a popup "Error: Resource not found":
Desired Solution you'd like
I'd like to just be able to to use Azure OpenAI for image generation. Just like for the chat models. Thank you!
Originally created by @holzerseb on GitHub (Aug 4, 2025).
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
### Problem Description
I am using Azure OpenAI for all models. With the GPT-image-1 and Dall-e-3 available on Azure OpenAI, we want to support image generation through azure. However, any attempts to configure the image generation with azure fails.
I have attempted to use the Azure OpenAI API URL instead of the default OpenAI URL and tested multiple options for the URL:
- https://[my_url].azure.com/openai/deployments/dall-e-3/images/generations?api-version=2024-02-01
- https://[my_url].azure.com/openai/deployments/dall-e-3/images/generations
- https://[my_url].azure.com/openai/deployments/
- https://[my_url].azure.com/openai/
- https://[my_url].azure.com/
I wanted to test it with Dall-e-3 first, so my default model is currently configured with "dall-e-3". Image size is 1024x1024. Steps is set to 1, but I've tried 10, 50, 100 too.
I'm using 4o for the chat.
While Open WebUI saves and enables the feature correctly, I get this response:
<img width="984" height="196" alt="Image" src="https://github.com/user-attachments/assets/020f8081-dda9-4f47-8db3-e6079d275043" />
Clicking on the "Generate image" button, results in a popup "Error: Resource not found":
<img width="2322" height="684" alt="Image" src="https://github.com/user-attachments/assets/11a52e94-4f6d-40b3-9470-d637996dfb91" />
### Desired Solution you'd like
I'd like to just be able to to use Azure OpenAI for image generation. Just like for the chat models. Thank you!
### Alternatives Considered
_No response_
### Additional Context
https://github.com/open-webui/open-webui/issues/15773
https://github.com/open-webui/open-webui/discussions/15774
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 @holzerseb on GitHub (Aug 4, 2025).
Check Existing Issues
Problem Description
I am using Azure OpenAI for all models. With the GPT-image-1 and Dall-e-3 available on Azure OpenAI, we want to support image generation through azure. However, any attempts to configure the image generation with azure fails.
I have attempted to use the Azure OpenAI API URL instead of the default OpenAI URL and tested multiple options for the URL:
I wanted to test it with Dall-e-3 first, so my default model is currently configured with "dall-e-3". Image size is 1024x1024. Steps is set to 1, but I've tried 10, 50, 100 too.
I'm using 4o for the chat.
While Open WebUI saves and enables the feature correctly, I get this response:
Clicking on the "Generate image" button, results in a popup "Error: Resource not found":
Desired Solution you'd like
I'd like to just be able to to use Azure OpenAI for image generation. Just like for the chat models. Thank you!
Alternatives Considered
No response
Additional Context
https://github.com/open-webui/open-webui/issues/15773
https://github.com/open-webui/open-webui/discussions/15774
@tjbck commented on GitHub (Aug 4, 2025):
PR welcome!