I trained this model using Kohya SS running locally on my PC, I used 20 high resolution images of model Nicola Cavanis. I recommend JuggernautXL <a target="_blank" rel="ugc" href="https://civarchive.com/models/133005?modelVersionId=240840">https://civarchive.com/models/133005?modelVersionId=240840</a> but it works fine on any of the most common SDXL models, even the less realistic, more stylized ones.20-30 steps are reccomended, no particular negative prompt is needed, use it at 0.75-1.00 strength

Nicola_Cavanis_Lora_SDXL

v1. Trained on Kohya SS on 20 images, 20 repeats reg images were used.

Files with SHA256: 0c956c986c76df0cad5b12631c580af269820de0ede2391c47d3c2be953c9242