ShengShu launches Vidu Q3 Reference-to-Video
ShengShu Technology has released Vidu Q3 Reference-to-Video globally, a feature that builds story-driven clips from several reference inputs at once, as Chinese video-generation labs compete on creative control and consistency.
ShengShu Technology announced the global launch of Vidu Q3 Reference-to-Video, a capability for story-driven video generation built on reference inputs.
According to the company, creators can reference and combine subjects, environments, costumes, props and visual styles in a single workflow. ShengShu says this improves creative control, consistency across shots and production efficiency.
The release also adds visual and audio features. On the visual side, ShengShu lists support for six types of cinematic effects: particle systems, fluid simulation, dynamic motion, camera movement, transitions and lighting. On the audio side, it lists five categories of sound generation: ambient sound, motion-driven audio, atmospheric layers, foley effects and emotion-driven cues.
ShengShu is targeting short-form series, animation, film and television, advertising and e-commerce, for both individual creators and enterprises.
The only performance evidence the company cites is a third-party benchmark. It says Vidu Q3 ranked No. 1 on SuperCLUE's first global Reference-to-Video leaderboard. The announcement does not give scores, the list of competing models or details of how the leaderboard was evaluated. Because the leaderboard is new and narrow, the result is hard to put in context.
The announcement does not cover pricing, API availability, output resolution, clip length or how many references can be used per generation.
The release also includes a headline saying Alibaba led a RMB 2 billion Series B to support ShengShu's "general world model" development. It gives no further details about the round, such as other investors, valuation or timing.