LORA trained on 50 publicly available images of Olivia Wilde.

Trained on base SD 1.5, should work with most models.

Use ADetailer if details are lost and strength of 1 should work on most prompts.
Consider leaving a review if you like the model. All feedback is appreciated.