2 Comments
User's avatar
Barry Phipps's avatar

Love it. I get the double person "mistake" often as well (more so with Stable Diffusion than with Flux). It happens most when the latent image size is made larger than the pixel dimensions the model was trained on (1080 pixels). Embrace the flaws of AI!

Fred Ritchin's avatar

It’s a mistake and, perhaps, a commentary. The flaws, we agree, can be quite revealing!