I created this model using 800+ images of the Gorgeous Ai Shinozaki
It was trained on v1-5-pruned for 5000 steps
In all the images I posted I was using Dynamic Thresholding at 3.5
It works well on dreamshaper_4BakedVae
Initial Version