Trained on Realistic Vision V5
Suggested Weight: around 0.8-1.0
This version is better than the last one because of image qualities and can consistently generate good teeth.
If you like this model, please consider leaving a like. Or you can buy me my next meal here. Thanks.❤️
Description
After a lot of testing, this version can consistently generate good teeth, and image qualities are better than the last one
FAQ
Comments (2)
tip: don't add too much detail. it may make the whole face look creepier, but can also give you some super lame results.
true, because what most 'add details' prompts do are to strengthen perfections (remove imperfections) and that's not what we want when drawing non-typical characters. that's why I remove many negative prompts when I try to draw this kind of characters. Also part of the reasons being that high quality images in my dataset are actually from a sculpture, close-up images may appear less realistic.