• Modern Chaos
  • Posts
  • Introducing Vidu, China's Answer to OpenAI Sora

Introducing Vidu, China's Answer to OpenAI Sora

On Saturday, the Chinese tech company ShengShu-AI and Tsinghua University launched Vidu, a new AI that turns text into video. This is the first AI text-to-video in China said to be as good as Sora.

Vidu is China's first advanced video model capable of producing long, consistent, and dynamic videos. His Co-founder Zhu Jun highlighted that Vidu represents a significant step forward in autonomous innovation, with capabilities that include simulating the real world, understanding complex camera interactions, and creating videos with a single click.

In January this year, the team achieved the generation of 4-second videos, which could match the effects of Pika and Runway. By the end of March, they had broken through to 8 seconds. In April, the team further intensified their efforts. Today, Vidu is showcased with a 16-second achievement, but Zhu Jun believes that in the near future, Vidu will iterate at an even faster pace.

Shengshu-AI, valued at $100 million and backed by major Chinese tech firms, has not yet released Vidu publicly.

My point of view:
Vidu isn't as good as Sora. It only makes 16-second videos, while OpenAI can make up to 60-second videos that look better. Congrats to Vidu on achieving significant results with a budget and compute power far from OpenAI's capabilities. The race has begun.

Reply

or to participate.