✦
Summary
Vidu is an AI video generator developed by ShengShu-AI and Tsinghua University, first unveiled at the Zhongguancun Forum in Beijing. Built on the self-developed Universal Vision Transformer (U-ViT) architecture, it transforms text, images and reference videos into high-quality video, and was positioned as China's first model on par with OpenAI's Sora.
Key Features
Text-to-video generation from written prompts
Image-to-video generation with first and last frame control
Reference-to-video generation combining multiple reference images into one video
Saved character, prop and scene references for reuse (My References)
Storyboard tools for planning multi-scene videos
High-quality video output up to 1080p resolution
Vidu Agent for guided creation from a single sentence to a complete work
Vidu App for collaborative creation on mobile
Free credits for new users to generate videos at no cost
About Vidu
The platform supports text-to-video, image-to-video and reference-to-video creation modes, with features for maintaining character and scene consistency, storyboarding multi-scene videos, and producing 2D animation and cinematic content. It simulates realistic physical motion, light and shadow, and detailed facial expressions, and generates videos up to 1080p with fast generation speed.
Vidu is used by independent creators, film and animation studios, and businesses for social media content, marketing videos and commercial advertising, with flexible individual and enterprise plans and a set of free credits for new users. It is suited to anyone who wants to turn ideas into video without traditional production complexity.