Finally I am sufficiently satisfied with the results to release a version of this LoRA! The goal is to produce the most faithful version of Lae'zel possible. Version updates are likely to come as I reach new breakthroughs in my skill and the model's fidelity, or when I can't figure out how to improve it any further.
Trained on hundreds of in-game screenshots. WD14 tagging with light edits.
I am new to stable diffusion and training LoRA and would appreciate any tips or advice to improve my skills!