Vidu is an AI video generation platform developed by ShengShu Technology, a Beijing-based company founded in March 2023 by a team from Tsinghua University's Institute for AI, with Tsinghua professor Zhu Jun as chief scientist. The platform was developed jointly with Tsinghua University and is backed by investors including Baidu and Alibaba.
Vidu supports text-to-video, image-to-video with first and last frame motion control, and a signature reference-to-video feature that uses up to seven reference images to keep characters and subjects consistent across generations. It also includes an AI sound effect generator for synchronized dialogue and effects, plus AI image generation. Output reaches 1080p with clips up to about 8 seconds, extendable further, and the platform is noted for its strength in anime-style generation and fast generation speeds.
The underlying models, Vidu Q1 and Q2, trace their architecture to ShengShu and Tsinghua's U-ViT diffusion-transformer research; the TurboDiffusion framework introduced in late 2025 accelerates generation substantially.
Vidu is available as a web app, mobile apps, and a developer API at platform.vidu.com. A free tier provides monthly credits plus unlimited generation in Off-Peak Mode. Paid plans start at around $10/month (Standard, 800 credits) and scale through Premium (~$35/month) and Ultimate (~$99/month) tiers, with enterprise options available.