My first SDXL1.0 LoRA. Trained with the same images from my Black Dress LoRA for SSD1.5
All images used the refiner (0.2 to 0.4 denoising strength) and sometimes restore faces)
Works best with using 1girl in prompt
Trigger word: blackdresslora
Trained on SDXL1.0
In case you want to support me or want a request/commission: Link