Lora

Details

Download Files (1)

Model description

Trained on a 26-image SDXL 1024x1024 dataset with detailed portraits, full-body poses and high-variety lighting. Best results at LoRA strength 0.6–0.9

Images made by this model