Vidu Q4 Preview is the first public preview of Shengshu Technology's next-generation flagship audio-video model. On October 7, 2026, Shengshu officially launched Vidu Q4 Preview through a company release, and on October 8 the model was also announced inside the YouCam app family. It is available through Vidu's web product and API platform, with launch pricing starting at $0.014 per second.
This upgrade is mainly about performance
According to the company, Q4 Preview improves character performance, camera work and visual effects: facial expression, emotion, body movement and voice are better coordinated, aiming to let characters convey feeling even in silent pauses. Its reference capabilities are the hard specs of this release — up to 15 reference images to build complex scenes and characters, and up to 3 reference audio clips to guide a character's voice and emotional delivery. Output reaches 2K or 4K resolution with 10-bit color depth, targeting material standards for professional workflows.
$0.014 per second is a bet on iteration costs
Video generation rarely succeeds in one take: a usable shot usually needs repeated adjustments to expressions, camera angles and openings. Shengshu co-founder and CEO Yihang Luo's message in the release was clear — efficiency gains should ultimately become creators' freedom to try one more version. Pricing from $0.014 per second targets the trial-and-error budgets of independent creators, small studios and high-volume teams in short drama and advertising, rather than only chasing the ceiling of a single demo clip.
Two things to watch
First, Q4 is still a preview: the company has shared capability directions and early creator feedback, but no public third-party benchmark comparison yet, so real-world stability will only be clear once more teams use it. Second, its strategy is plain to see — attack the industry's character-consistency problem with multi-image and audio references, then widen adoption with a low price. If the full release holds both consistency and price, short drama and advertising, the workflows most sensitive to characters changing face between shots, will feel the change first.
