MeiYi-dev

I don't know where to ask this, but does your lora trainer support every model that ComfyUI supports? Also, can you please include optimizers from this project https://pytorch-optimizers.readthedocs.io/en/latest/? (some optimizers...

> > I don't know where to ask this, but does your lora trainer support every model that ComfyUI supports? Also, can you please include optimizers from this project https://pytorch-optimizers.readthedocs.io/en/latest/?...

Soon it will get native ComfyUI support. The model is much better in motion dynamics and sematics than Wan2.2 and it's a single model unlike Wan so it will be...

Yeah it's now natively supported in ComfyUI and diffusers. Can we get a distlled model please.

> [@MeiYi-dev](https://github.com/MeiYi-dev) are LoghtX2V Loras obtained via RCM? There are multiple distillation methods for Wan2.1 on HuggingFace. 1: [Causvid](https://causvid.github.io/): https://huggingface.co/lightx2v/Wan2.1-T2V-14B-CausVid 2: [Self Forcing](https://self-forcing.github.io/): https://huggingface.co/lightx2v/Wan2.1-T2V-14B-StepDistill-CfgDistill-Lightx2v 3: [DCM](https://vchitect.github.io/DCM/): https://huggingface.co/cszy98/DCM 4: [AccVideo](https://aejion.github.io/accvideo/): https://huggingface.co/aejion/AccVideo-WanX-T2V-14B...

> [@MeiYi-dev](https://github.com/MeiYi-dev) Fast Wan is Distribution Matching (DMD2) I see that Kandinsky 5 Pro is going to get a 16 step distilled version. Can we also get a 1-4 step...