FastVideo releases 4-step distilled text-to-video model on Hugging Face

FastVideo · hf · 2026-08-29

A trending model named FastVideo-FastH3-4-step-Preview-v1-VSA-DataFree has appeared on Hugging Face. This is a text-to-video pipeline utilizing distillation and DMD2 methods, capable of generating videos in just a few steps. It also supports text-to-audio-video generation.

Original post →

More from Multimodal

Multimodal channel →