What dataset was Kroma trained on?
I'm mostly curious how many distinct concepts from places like Danbooru/e621 are in there, aside from just photorealistic photos.
Best approach is to lurk on lodestones datasets like https://huggingface.co/datasets/lodestones/booru-essence or https://huggingface.co/datasets/lodestones/e621-captions . Since they exist , they are most likely used to train Krea 2.
For my own usage I compiled the parquets listed in the repo links above into perchance text generators (you can cycle between the sorted categories using prev gen / next gen , 68 generators in total , and probably ~40K prompts in total)
https://perchance.org/fusion-t2i-essence-solo-true-multi-false-general-cluster-001-1
Code for creating the generators https://huggingface.co/datasets/codeShare/lora-training-data/blob/main/sort_parquets_by_category.ipynb
Cheers