Vidu - AI Video Generation Platform

Overview

Vidu is China's first long-duration, high-consistency, high-dynamic video foundation model, launched by Beijing Shengshu Technology Co., Ltd. in collaboration with Tsinghua University. Founded in 2023 and led by Tsinghua professor and IEEE Fellow Zhu Jun, the company has rapidly emerged as a leader in AI video generation. Vidu supports text-to-video, image-to-video, and reference-to-video creation modes, designed for independent creators and teams.

Core Product Lines

Vidu Q3 / Q3 Pro

Launched in January 2026, Vidu Q3 scored 1241 points on the international benchmark Artificial Analysis, ranking #1 in China and #2 globally — behind only xAI's Grok and ahead of Runway Gen-4.5, Google Veo 3.1, and OpenAI Sora 2. It supports up to 16-second 1080P video with synchronized audio, the world's first model to achieve high-quality audio-video generation at this duration. Features include multi-shot switching, multilingual dialogue, text rendering, and reference-to-video capabilities.

Vidu Q2

Centered on the breakthrough of "AI acting," Vidu Q2 achieves unprecedented micro-expression generation, crossing the chasm from "generating video" to "generating performance." It understands and renders subtle facial micro-expressions, enabling digital characters to deliver emotionally compelling performances. Supports first-and-last-frame video, cinematic and flash generation modes, with 2-8 second duration options.

Vidu Claw

The latest AI video application product that rapidly generates advertising video concepts and finished clips from a single sentence description, around product selling points, communication goals, and content style.

Technical Highlights

• Universal Reference: Maintains character consistency and scene continuity through multi-type element referencing (subjects, scenes, costumes, props)
• Six Special Effects: Particles, fluids, dynamics, camera motion, transitions, lighting
• Five Sound Effects: Environment, motion, atmosphere, foley, emotion
• Fast Generation: ~5 minutes for 10s 1080P, ~1 minute for 10s 720P
• 1080P Output: Pricing as low as 0.2 RMB/second

Market Performance

Vidu serves users across 200+ countries and regions through MaaS (Vidu AI Open Platform) and SaaS (Vidu Agent, Vidu Claw), covering interactive entertainment, advertising, anime/film, and cultural tourism. In April 2026, Alibaba Cloud led a Series B round valuing the company at 12 billion RMB, and the company is preparing for an IPO. Vidu is also available on Alibaba Cloud's Bailian platform.

Creation Modes

• Text-to-Video: Generate high-quality videos from text descriptions
• Image-to-Video: Transform static images into dynamic videos
• Reference-to-Video: Upload reference videos to guide content generation while maintaining subject consistency