First attempt of making a Lora
Trained with 1.5 base model @~15k steps / ~100 Images (512x512)
All example images are generated with RealisticVisionV2 txt2img
vlry.safetensors