Tablet menu

What does the most beautiful place in the world, according to the Neural network, look like?


Choosing a neural network and compiling a query

In one of the chats of the project in which I participate, one of the participants shared a link to the Kandinsky 2.1 neural network, which generates images. Generates by text, by image, and you can also "mix" text and image.



I liked it, to be honest. At first I dabbled, made some crazy requests, then, of course, I added my photo with different descriptions.

Further, I was interested in what the neural network thinks about beauty. I decided to start with nature, namely the request - "Noisy, beautiful river", and then completely - "The most beautiful place in the world".

See what happened.

Difficulties with using a neural network

I used "Kandinsky" in the telegram bot, there were no problems. Everything is clear and pretty fast.

Last modified on
back to top