Any-Step Video Diffusion Model with On-Policy Flow Map Distillation
Create short AI videos from text or first and last frames, with synchronized dialogue, effects, and ambience.