Native universal video foundation model by ShengShu and Tsinghua University, generating 1080P cinematic clips in seconds.
Initial free generation credits upon registration, supporting 1080P ultra-fast video rendering
Vidu is a native video foundation model built on the Universal Vision Transformer (U-ViT) architecture, co-developed by ShengShu Technology and Tsinghua University. Its defining superpower is unmatched generation velocity coupled with pristine 1080P resolution. Generating up to 8-second clips in tens of seconds, it effortlessly simulates complex real-world physics: liquid splashes, rolling smoke, optical depth of field, and coordinated multi-subject movements. Featuring cinematic camera pathing presets, Vidu acts as an industrial-grade visual production engine for films, animations, and marketing agencies.