Gallery

Welcome. Here, you will find several sets of images representing the typical outputs of each of the three image generation models at the center of this project, Stable Diffusion 1.5, Stable Diffusion XL, and Flux.1.

Along with each image, you'll see the text prompt that was used to generate it, drawn directly from analyses produced by classifier models. I have made no effort either to correct or editorialize these prompts or the images. You're seeing everything exactly as the AI models intended.

For some of the subjects, I have also created composite images that illustrate the average face these models tend to generate. These have been lightly color-corrected to improve visibility.

So, what prompts do you think produced the images below? Select any one of them to find out.