GoWithTheFlow Hunyuan LoRAs for Diffusers

[tin2tin]

https://huggingface.co/spacepxl/HunyuanVideo-GoWithTheFlow-unofficial/tree/main/diffusers

So how do we use them if we have no vid2vid for Hunyuan?

[a-r-r-o-w]

We are bit behind on editing methods for vid2vid to work. The following methods are on our radar:

• RF Inversion: https://rf-inversion.github.io/
• FlowEdit: https://github.com/fallenshock/FlowEdit
• Tokenflow: https://diffusion-tokenflow.github.io/
• Fireflow: https://github.com/HolmesShuan/FireFlow-Fast-Inversion-of-Rectified-Flow-for-Image-Semantic-Editing

There are no plans of creating additional pipelines to support these methods as of now. Many of these works build upon diffusers and have pipeline examples that may be adapted to other models (for the moment), but we will definitely support it in #9672 eventually