Trained from 20 images in OVA video.
Clothes and body shape are not trained, sorry.
SDXL_Sylvia_Nimrod_v1.safetensors