Flux FIll lora / dreambooth finetune support

[sanbuphy]

I would like to ask if there are plans to support LoRA fine-tuning for Flux Fill (inpainting)? I see that the community has already supported related code, and I would like to ask if this can be directly integrated? Thank you
https://github.com/huggingface/diffusers/blob/7a350cc8da60ee67f12a5bf5d3b3fcb4c06ffdb0/examples/research_projects/dreambooth_inpaint/train_dreambooth_inpaint_lora_flux.py

My request is to add support for DreamBooth fine-tuning and LoRA fine-tuning with flux fill.
Currently, the official implementation does not have this feature.
I can help promote integrating this into the official version, including support for loading and fine-tuning models in FP8.

[murmurelo111]

May I ask which community provided this code?

[yiyixuxu]

hi @sanbuphy feel free to send e PR!
the training code should go in this folder https://github.com/huggingface/diffusers/tree/main/examples

cc @sayakpaul @a-r-r-o-w

[sayakpaul]

Oh that would be awesome. I think we could start with LoRA and see some results first. Please feel free to start the PR and we can help as needed :)

[sanbuphy]

hi @sanbuphy feel free to send e PR! the training code should go in this folder https://github.com/huggingface/diffusers/tree/main/examples

cc @sayakpaul @a-r-r-o-w

ok! is that correct ? https://github.com/huggingface/diffusers/pull/11393/files

[sanbuphy]

Oh that would be awesome. I think we could start with LoRA and see some results first. Please feel free to start the PR and we can help as needed :)

ok, thanks ! https://github.com/huggingface/diffusers/pull/11393/files

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.