Shengshu Technology Launches Vidu Q4 Preview with 4K Video Output

AI video generation tools keep evolving, with multi-modal references and high-resolution output becoming key focus areas.

Shengshu Technology has released Vidu Q4 Preview, a new flagship video generation model. It supports image-to-video and reference-to-video modes, creating videos from images, text, multiple reference images, and audio.

The model allows up to 3 audio clips and 15 reference images, keeping character voices consistent and giving full visual references for people, scenes, and props. It also improves camera movement and shot transitions, supports 2K and 4K output with 10-bit color depth, and generates up to 16 seconds per clip.

Vidu Q4 Preview starts at 0.09 yuan per second, covering specs from 540P to 4K. It suits AI short dramas, ads, and film production.