This Lora was trained on 66 digital art pieces.

I used WD14 for captioning so using 1girl/1boy might help

I used the default civitai traing which uses the SDXL Base model, I think but it also works great with other models like Juggernaut XL or AAM XL. If you want me to train the model on a specific base please just write a comment or write me directly via the chat.

PS: If you want the dataset please contact me. I just don't want to get CivitAI in copyright trouble.