ComfyUI [New Model] LongCat-Video (with weights!)

Feature Idea

LongCat-Video is a new video generation foundational model with 13.6B parameters from meituan-longcat, who have just released a LLM as well.

It differs from other models, such as Wan, in native Block Sparse Attention (practically like the recent HoloCine (add it too please :))), enabling it to generate minutes-long video without much computational cost and quality degradation.

The model is free, opensource and weights-available under MIT License.

https://huggingface.co/meituan-longcat/LongCat-Video

Existing Solutions

https://github.com/meituan-longcat/LongCat-Video -- official repo and inference code.

Other

Implementing Block Sparse Attention will also enable integrating the HoloCine Wan-derived model to produce Sora2-like videos with cinematic transitions. HoloCine repo: https://github.com/yihao-meng/HoloCine.

Oct 25 '25 12:10 kabachuha

Looking forward to implementation

Oct 26 '25 04:10 Pythonpa

https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/longcat

Kijai implementation in Wan Wrapper

ComfyUI format models:

https://huggingface.co/Kijai/LongCat-Video_comfy

Oct 27 '25 08:10 kabachuha

Would love to see it implemented natively in ComfyUI. It is promising.

Oct 28 '25 06:10 Pandaabear0

Any idea if it will be implemented in ComfyUI? Thanks.

Oct 28 '25 18:10 dvision1979

Looking forward to implementation

Oct 29 '25 09:10 xiaopengyou123

https://github.com/kijai/ComfyUI-WanVideoWrapper/tree/longcat

Kijai implementation in Wan Wrapper

ComfyUI format models:

https://huggingface.co/Kijai/LongCat-Video_comfy

Works fantastic, Just needs the refinement (480p 15fps to 720p 30fps) stage

Oct 30 '25 00:10 CCpt5

Any word on if this is something you guys are considering at least?

Nov 03 '25 16:11 Pandaabear0