This model is the result of a partial NVFP4 quantization of Wan2.1-I2V-14B-480P-StepDistill-CfgDistill-Lightx2v by lightx2v, produced using convert_to_quant by silveroxides. Some layers have been kept on their original BF16 format, while others were quantized as MXFP8 or NVFP4, mostly.
Wan2.1-I2V-14B-480P-StepDistill-CfgDistill-Lightx2v is an image-to-video generation model built on Wan2.1-I2V-14B-480P. It applies step distillation and classifier-free guidance distillation to reduce inference to 4 steps without CFG, cutting generation time substantially while preserving output quality.
IMPORTANT
Since NVFP4 is only supported on NVIDIA Blackwell architecture GPUs, running this model requires a Blackwell GPU with its corresponding support enabled in torch, along with a recent version of ComfyUI and comfy-kitchen built against CUDA 13. You'll also need at least torch 2.10 to make it run, so this is not for the faint of heart with all the corresponding, if you know what I mean.
The model can be used in ComfyUI with the following parameters, based on the distilled model's own recommendations:
Shift: 5.0
Sampler/Scheduler: euler/simple
CFG: 1.0
Steps: 4
Other combinations are possible, such as lcm, heun/linear_quadratic, etc.