Since the model is trained exclusively on images from the movie, will not perform as well in other situations.

LoRA trained on 14 screencaps of Sophia Lillis in the Dungeons and Dragons movie.

Trained on PonyXL, should work with other pony based checkpoints.

Use ADetailer if details are lost and strength of 0.7-1 should work on most prompts.

Trained for 1, 2 repeats (6, 8), 50 epochs, network 16/2.
Consider leaving a review if you like the model. All feedback is appreciated.