I have searched the existing issues and discussions.
Problem Description
Currently, when asking for an edit while working with an image, follow up images are not edits if asked for, but new pictures which roughly fit previously described generation tasks. There is an different openai endpoint for editing (https://platform.openai.com/docs/api-reference/images/createEdit) that could be used, it requires a couple of things though (Must be a valid PNG file, less than 4MB, and square. If mask is not provided, image must have transparency).
Desired Solution you'd like
When asking for an edit of a picture, it will not call the generic dall e endpoint, but instead check if the image fits the criteria (4MB, png format) and if it is not to spec, convert it to this spec. It should then call the editing endpoint and return the edited image.
Alternatives Considered
No response
Additional Context
No response
Originally created by @LOFWagner on GitHub (Mar 31, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/12232
### Check Existing Issues
- [x] I have searched the existing issues and discussions.
### Problem Description
Currently, when asking for an edit while working with an image, follow up images are not edits if asked for, but new pictures which roughly fit previously described generation tasks. There is an different openai endpoint for editing (https://platform.openai.com/docs/api-reference/images/createEdit) that could be used, it requires a couple of things though (Must be a valid PNG file, less than 4MB, and square. If mask is not provided, image must have transparency).
### Desired Solution you'd like
When asking for an edit of a picture, it will not call the generic dall e endpoint, but instead check if the image fits the criteria (4MB, png format) and if it is not to spec, convert it to this spec. It should then call the editing endpoint and return the edited image.
### Alternatives Considered
_No response_
### Additional Context
_No response_
@LOFWagner since I came across the same question, could you please explain, did you found the plugin solution, or some workaround?
<!-- gh-comment-id:2773892931 -->
@smartynov commented on GitHub (Apr 2, 2025):
@LOFWagner since I came across the same question, could you please explain, did you found the plugin solution, or some workaround?
Hey, sadly I did not find a solution yet. I merely closed this ticket because the image generation feature in webui is from a community plugin and according to this Git Projects rules, you cant open tickets for support with plugins.
<!-- gh-comment-id:2774629559 -->
@LOFWagner commented on GitHub (Apr 3, 2025):
Hey, sadly I did not find a solution yet. I merely closed this ticket because the image generation feature in webui is from a community plugin and according to this Git Projects rules, you cant open tickets for support with plugins.
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 @LOFWagner on GitHub (Mar 31, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/12232
Check Existing Issues
Problem Description
Currently, when asking for an edit while working with an image, follow up images are not edits if asked for, but new pictures which roughly fit previously described generation tasks. There is an different openai endpoint for editing (https://platform.openai.com/docs/api-reference/images/createEdit) that could be used, it requires a couple of things though (Must be a valid PNG file, less than 4MB, and square. If mask is not provided, image must have transparency).
Desired Solution you'd like
When asking for an edit of a picture, it will not call the generic dall e endpoint, but instead check if the image fits the criteria (4MB, png format) and if it is not to spec, convert it to this spec. It should then call the editing endpoint and return the edited image.
Alternatives Considered
No response
Additional Context
No response
@LOFWagner commented on GitHub (Mar 31, 2025):
My bad, It is a community contribution, closing this
@smartynov commented on GitHub (Apr 2, 2025):
@LOFWagner since I came across the same question, could you please explain, did you found the plugin solution, or some workaround?
@LOFWagner commented on GitHub (Apr 3, 2025):
Hey, sadly I did not find a solution yet. I merely closed this ticket because the image generation feature in webui is from a community plugin and according to this Git Projects rules, you cant open tickets for support with plugins.