Trained on Illustrious v0.1
use: depth of field and blurry background for best results
Depth.safetensors