Base Pony tuned on an additional ~3500 images. Each image was captioned with both Florence-2 Large "More Detailed" captions and also Booru tags from WD-VIT-Tagger V3.
Description
Improved basically everything.
Base Pony tuned on an additional ~3500 images. Each image was captioned with both Florence-2 Large "More Detailed" captions and also Booru tags from WD-VIT-Tagger V3.