Vidu

Vidu is an AI video generation platform developed by Beijing Shengshu Technology Co., Ltd. The platform converts text prompts and static images into dynamic video content using a domestically developed multimodal large model (UniDiffuser). It supports three primary generation modes: reference video-to-video, image-to-video, and text-to-video generation, alongside a template library and specialized tools such as Vidu Claw and Vidu S1. The platform emphasizes fast generation speed, multi-subject consistency across frames, off-peak generation mode, and high-definition output with smooth animation quality. Users report generating creative video content in approximately three steps. Vidu operates on a credit-based system where subscription credits expire after 30 days while purchased credits remain available until used. The platform offers an API open platform for developers and serves users across more than 200 countries, with tens of millions of registered users. Security and privacy documentation is available, describing data collection practices for account registration (phone number and verification code), content generation (prompts, uploaded materials including facial feature points and contours, and generated output), payment processing, and device information for service stability and fraud prevention. Important coverage gaps exist: developer documentation, API pricing details, dedicated security pages, and trust/compliance pages returned errors or were not accessible during review. Teams evaluating Vidu for workflows requiring API integration, developer documentation, or detailed security certifications should contact the vendor directly at [email protected] to confirm current availability and specifications.