The American company Nvidia developed a neural network that was trained to generate video based on a text description.

At the heart of the neural network is the development of the Stable Diffusion neural network.

Video resolution is up to 1280×2048 pixels with a frequency of 24 frames/second and takes into account about 4.1 billion parameters, of which 2.7 billion are learned from the video.

Currently, this is only a research development, until we talk about the appearance of a new neural network in the public domain.