stable-diffusion.cpp T2v or i2v support

Can we please get animatediff or hunyuan , ltxv, cogvideo,sana ,skyreels support?

Mar 12 '25 19:03 KintCark

I'm actually building a tool using Imgui/SDL3/stablediffusioncpp and video models (especially wan2.1) support a feature I would love to add but I don't know much about 'around models' programming. Does someone knows if it's doable with the actual ggml library? If so I may try in a near future.

Mar 15 '25 16:03 ImKyra

I'm actually building a tool using Imgui/SDL3/stablediffusioncpp and video models (especially wan2.1) support a feature I would love to add but I don't know much about 'around models' programming. Does someone knows if it's doable with the actual ggml library? If so I may try in a near future.

It's harder than it should be because most video models use 5D tensors at some point, and GGML only supports up to 4D tensors. I don't think it's undoable, but it's definitely tricky to work around this limitation. I gave up implementing LTX because of that.

Mar 15 '25 17:03 stduhpf

Thanks for your insight. Yeah that seem to be a hard things to work around without patching the lib itself which seems also to be pretty hard.

Mar 15 '25 19:03 ImKyra

Support for the Wan video model has been added.

Sep 14 '25 14:09 leejet

Wan don't work on 7.5 go ram i even try the smallest quantization and get all the way to starting steps and it says I need 16.8mb more memory I was so close

Sep 14 '25 21:09 KintCark

the vae decode currently requires 20+ gigs or so.

Sep 14 '25 21:09 Green-Sky