English to Russian Tech/Engineering Computers: Systems, Networks

First, your text prompt gets projected into a latent vector space by the text encoder, which is simply a pretrained, frozen language model. Then that prompt vector is concatenate to a randomly generated noise patch, which is repeatedly "denoised" by the decoder over a series of "steps" (the more steps you run the clearer and nicer your image will be -- the default value is 50 steps).

https://keras.io/guides/keras_cv/generate_images_with_stable...