Chinese company Shengshu Technology introduced Vidu, a text-to-video creation tool developed with Tsinghua University. It is claimed that the model is very close to OpenAI’s Sora model, which is still unpublished.
Generative artificial intelligence models are no longer just photographs, they are indistinguishable from the real thing. videos too can create. The best example of this is “Sora”, introduced by OpenAI in recent months. This model, which has not yet been widely used, could create ultra-realistic videos from texts in a very short time.
Now, a model that can be a serious rival to OpenAI’s Sora has been introduced. “viduThis artificial intelligence tool, named “, came from China.
It can create 16-second 1080p resolution videos with a single click!
*If you can’t view the video link here Click.
China’s first text-to-image creation model Described as Vidu, it was introduced within the scope of the technology event Zhongguancun Forum held in Beijing today. Examples shared from the model revealed that the race for productive artificial intelligence will intensify not only in the USA but all over the world.
Vidu was developed in collaboration with Chinese artificial intelligence company Shengshu Technology and Tsinghua University. In the statement, it was stated that the model can be operated via text commands. 16 seconds long at 1080p resolution It was reported that he could create videos.
According to the company’s statement to China Daily, Vidu can both create videos for longer periods of time and offer a high level of consistency. Moreover He’s also very close to Sora’s level. It is claimed. When we look at the examples shared, we can really see this. He can create videos that look like real footage as successfully as Sora.
RELATED NEWS
OpenAI Shared Artists’ First Impressions of the “Sora” Model: New Videos That Will Leave Your Mouth Dropped Are Also Released
Shengshu Technology has developed the basic architecture of the model. From September 2022 He stated that it has been developed since. This means it has been in the works for longer than Sora. It is also reported that, unlike his rival, he can understand elements specific to Chinese culture, such as pandas and dragons. There is no information yet about when the model will be offered to users.