Stylization by Audrey Benjaminsen.

45 images were used to train the model.

the training of the LoRA was on the CyberRealistic 3.1 model.

Training with a resolution of 640x960 images.