Videos
Visual pages, now in motion.
The same visual-page interface transfers to video. HunyuanVideo-1.5 takes V2V-Zero's visual specification pages and produces short coherent clips—supplying complementary evidence that V2V is a general conditioning paradigm.
20.2
Video · Overall /100
616
Videos Generated
7
Task Categories
33
Frames @ 480p
Representative Examples
HunyuanVideo-1.5 on Simple-V2V Bench.
Figure 4.
HunyuanVideo-1.5 representative examples on Simple-V2V Bench. Each row shows the visual input page and four uniformly sampled frames from a generated video. The aggregate 20.2/100 is computed over 616 generated videos using a VLM judge on the full video.