LORA trained on 70 publicly available images of Scarlett Johansson.

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

Pony

This version was trained on (15, 11) images with (24, 16) repeats on Pony Realism.
The face likeness is almost there, but further training did not improve it.
I have included an image where I used facedetailer/inpainting for the face with my SD 1.5 model. This provides the best output IMO.
Also added a comparison of Pony facedetailer vs 1.5.