If you're like me, you've found yourself thinking "wow, why is it so hard to get StableDiffusion to produce unattractive facial hair??".

This is my first LoRA, and I'm reasonably happy with it. It's probably a bit overtrained on too few faces, but it produces the results I was looking for and I'm pretty happy about that!

It was trained on SDXL-1.0 base, but I've been testing with DreamShaperXL, both v8-Turbo and v9-Lightning, cfg 2.0, 6 steps, dpmpp_sde/karras, with LoRA strength at 1.0.

Hope you find it fun!