Can nunchaku support more Video-generator model？

[philipwan]

In addition to flux, does nunchaku support prompt-to-video models, such as OpenSoraPlan, Cogvideo，etc.
Flux is also a transformer-based model

[lmxyy]

We are looking into this and will definitely support video generation models like Mochi and CogVideoX. Stay tuned.

[nitinmukesh]

That's good news.

[philipwan]

We are looking into this and will definitely support video generation models like Mochi and CogVideoX. Stay tuned.

Wow~Good News!!!
I notice there are too many new operators in nunchaku/src, Aren't these universal operators for other model？If it is to adapt to different models, what else does nunchaku need to do? I also want to try it

[lmxyy]

I think we just need to quantize the model with deepcompressor and then convert it to nunchaku format.
Yeap, we have all the operators in the C++ files. We are still cleaning the code to support modularization. With that, I think it is easy to extend nunchaku to new models.

[philipwan]

I think we just need to quantize the model with deepcompressor and then convert it to nunchaku format. Yeap, we have all the operators in the C++ files. We are still cleaning the code to support modularization. With that, I think it is easy to extend nunchaku to new models.

Thanks！

[marvin-0042]

We are looking into this and will definitely support video generation models like Mochi and CogVideoX. Stay tuned.

awesome! thank you so much! Do you have rough, high level timeline when this will support Mochi 1? Thanks!