Shengshu releases Vidu Q4 Preview video model with 4K output, multi-image and audio references
Shengshu Technology has launched Vidu Q4 Preview, a preview of its flagship video generation model, supporting image-to-video and reference-to-video with up to 15 reference images and 3 reference audio clips, 2K/4K output at 10-bit colour, and clips of up to 16 seconds. The company quotes a launch price from ¥0.09 per second across 540P–4K tiers, and says image-to-video runs about twice as fast as comparable models, with a 5-second clip produced in as little as 110 seconds. A hands-on review found the model stable in facial emotion, camera choreography and multi-object effects, though the final version is not yet out.
If the low price and speed gains hold in the full release, it could shape whether ad, short-drama, anime and e-commerce teams put AI video into regular budgets.