SDXL LoRA of Winona Ryder early 1990s era. The model was trained using Dreambooth(~fine tuning) using HD+ still frames (Shotdeck) mostly from "Reality Bites", along w/ some HD media photos of her from around the time. After the model was trained via Dreambooth/Finetuning this LoRA was extracted using Kohya scripts.

Trigger with "Winona Person" - (Yes you should include "person")

Trained on SDXL base 1.0 so likeness will come out best if you use that model. It should work relatively well with models trained using SDXL v1.0. **Play with weighting for flexibility. (Re: change the LoRA weight in the <> code to .8, .7, .6 etc if you needed to get more creative control, or try raising it to 1.2, 1.4, etc if the likeness seems "weak").