Alibaba has released its video and video production model to the public

Chinese holding Alibaba Wednesday announced the Model of Artificial Intelligence Production of its Image and Video by Name WAN 2.1 Made public and open to users. This can increase the use of this model and intensify competition in the field of artificial intelligence.

According to Reuters, Alibaba has announced four copies of the WAN 2.1 model: T2V-1.3B T2V-14B, I2V-14B-720 and I2V-14B-480p. These models can produce new images and videos through text and video inputs. The number 14b in the names of the models indicates the use of 14 billion parameters, which allows for more data processing and more accurate results.

These models are worldly available through Alibaba Claude and Hugging Face platforms for research, academic and commercial purposes.

Alibaba announced the news after a similar news from Deepseek Startup. This startup attracted the attention of technology investors by offering open source and low -cost models, and with a considerable performance with companies like Openai, it surprised this costly industry.

Artificial Intelligence Production of Video and Image Alibaba

Alibaba introduced the new version of his video and image production model in January and named it from WANX To WAN Changed. The company has announced its ability to produce very realistic images as one of its strengths.

Alibaba has also announced that the model has gained superiority in the VBENCH rating, which is a platform for evaluating video production models, and has performed better in capabilities such as interacting with multiple objects in a video than competitors.

In addition, the Chinese giant announced this week that in the next three years, at least 380 billion yuan ($ 52 billion) will invest in cloud computing and artificial intelligence infrastructure.