Comment by minimaxir
1 day ago
Some examples of the Reddit posts/memes:
https://reddit.com/r/StableDiffusion/comments/1x79gth/why/
https://reddit.com/r/StableDiffusion/comments/1x7ktbu/enshit...
I gave Flux 3 Klein my typical human test that has never failed on newer image models and it did fail; in addition to lower quality, the fingers were definitely not the correct proportions.
Whether the cause is due to more safeguards against NSFW content is hard to confirm technically; it was more an intuition back in the Stable Diffusion XL days but I suspect modern training paradigms are better at correlative inference.
i think its instructive how the ancient sculptors and artists used to have to draw nudes in order to understand the human body. perhaps our image models need this
It isn't just the ancient sculptors and artists (unless 20 years ago is ancient :-P), when i went to an art school there were nude drawing lessons there too.
Does it actually help? What about just drawing censored versions or those covered up with a leaf kind? I'm not trying to be difficult but just curious how the whole thing works. Like I don't think drawing clothed models just makes you not understand human anatomy. Also totally understand if it's just some tradition thing people so or whatever.
did you try with https://huggingface.co/black-forest-labs/flux-3-image-klein-... apparently the model is not going to work well unless this type of "prompt upsampler" is added
The example image for that upsampler shows a man holding a leash with all six of his fingers.
https://github.com/black-forest-labs/flux-3-image-klein/blob...
By the veins, this guy has an extra pointer finger, not the expected extra pinkie!
I did it in OpenRouter, which hits the direct BFL API.