I've trained this LoRA on my images from Tensor.Art. Suggested partner LoRA is J.Civet Graphic Novel Style with a very low slider control (0.10) for textured brush strokes. There is no benefit to complex prompts, as the LoRA only knows a few beautiful women for now. It suffers from most of the SD 1.5 flaws, like bad faces.
Run the image back through ADetailer with these settings if needed:
ADetailer model: face_yolov8m.pt,
ADetailer prompt: Highly detailed face with straight, beautiful nose, clear, beautiful eyes, and smooth skin.
ADetailer negative prompt:
ADetailer confidence: 0.5
ADetailer dilate/erode: 4
ADetailer mask blur: 4
ADetailer denoising strength: 0.6
ADetailer inpaint only masked: true
ADetailer inpaint padding: 32
