Hey @Sitron! So I have been trying to use this LoRA in different environments, and I just don't get the same results as you. Not even close, there just isn't enough resemblance. Can you share your setup as precisely as possible? Are you using a1111? How many generations do you need to get good results?

Here is what I have tried:

1. DrawThings app on macOS: Produces ok results, but there just isn't enough resemblance.
2. Easy Diffusion UI on macOS: This LoRA just straight up doesn't work using this UI. It always outputs a black image at strength 1. And when turning down the strength to 0.01 it outputs an image, but there is barely any resemblance.
3. a1111 on Google collab: This is where the LoRA works "the best" but It's still not giving me enough resemblance like in your images, even when using the same seed, euler a, 30 steps, 7 CFG.