This model is based on SDXL and is designed for creating photographic-style portraits (primarily headshots). The training set images are all 512x768 in size. The style references the work of Yoshiyuki Okuyama and similar Instagram-style portrait photography. Testing has shown that the overall results are quite satisfactory.
It is necessary to use FaceDetailer to assist with facial refinement.
Please put the style trigger words first and increase the LoRA weight to get the ideal effect