same problem, generating images from text, for example, although, in what are we cooler, although kondinskyell, that is, already, as if already, we recently presented a model that generates videos, called kandinsky video, although probably it should have been called something like tarkovsky, well, we called kandinsky video, we will continue this line, that's it, but what does better, knows the domestic cultural code better, of course, this is what we actually set for the model, what task, and we will certainly develop this further, what is the problem with russian data , here is knowledge of the russian domain, in that there is russian data, of course, there is less of it in what the model learns on, it learns on data from the internet, that is, on open image pairs. description, it kind of captures data at the time of the request, no, naturally, there is a team that deals with data, and that is, there is data it already contains, it does not search the internet at this moment, no, not by any means. of course, the data development team is responsible for this and this is a super-important ac