China unveils Vidu, Sora-level text-to-video large model

Summary: China unveiled Vidu, a Sora-level text-to-video large model, at the 2024 Zhongguancun Forum. Developed by Shengshu Technology and Tsinghua University, Vidu can create 16-second, 1080p videos in one click. It's China's first video model with extended duration and dynamic capabilities, close to Sora's level. Vidu can understand and generate Chinese elements like pandas and dragons. Its core architecture was initiated in September 2022.

Full article