Alexander Vedyakhin, head of Sberbank, announced this at the AI Journey conference. The model generates video with a resolution of 512×512 pixels, lasting up to eight seconds and at a rate of 30 frames per second. The video generation time is up to three minutes.
Kandinsky Video includes two main blocks. The first block is responsible for creating keyframes that define the structure of the video story. The second block generates interpolation frames, ensuring smooth motion in the final video. Both blocks are based on the Kandinsky 3.0 model of image synthesis from textual descriptions.
Alexander Vedyakhin, First Deputy Chairman of the Management Board of Sberbank, announced a new model developed by the company that allows generating full-fledged videos based on textual description. This is a significant step forward in the development of generative neural networks in Russia. Users will have even more opportunities for creativity and realization of their ideas in video format. Creation of unique videos will be available absolutely free of charge. The model, similar to other Sberbank products, will be available in the public domain. The company believes that artificial intelligence will open up new opportunities for people and provide tools for solving various tasks.
Ailib neural network catalog. All information is taken from public sources.
Advertising and Placement: [email protected] or t.me/fozzepe
English 
Русский