-
Notifications
You must be signed in to change notification settings - Fork 6k
[Feature request] ostris / Flex.2-preview. Open Source 8B parameter T2I Diffusion Model with universal control and inpainting support built in. #11394
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
Really cool model but this looks like it'll require a new pipeline implementation to accept canny/depth/openpose and other inputs for concatenation with latents |
Thanks for the info, @a-r-r-o-w . Please could you add New pipeline/ Community contribution tags, if someone wants to give it a go. |
@nitinmukesh Any reason for closing this FR? We would love to support it once someone on the team has the bandwidth, but until then I think this might be nice for good first contributions, no? |
My apologies, somehow my comment was not added while closing the issue. It is working. The author provided custom_pipeline for diffusers. I tested it. It's in Model card. |
Oh, that's amazing! But it uses a remote pipeline, and I think it'll be an impactful model to support. Is it okay to still keep this issue open so that we can track natively adding support for it? |
Oh absolutely. I reopened it. |
It was not a bug so removed unnecessary information.
The next version of Flex.1-alpha is released.
https://huggingface.co/ostris/Flex.2-preview
The solution is provided using custom pipeline by author of model which is working. Please refer the Model card for code.
However, it will be good to have this integrated in diffusers as core function.
In future other variants of Flux schnell may appear with similar functionality.
The text was updated successfully, but these errors were encountered: