SDXL

Also try the Pony and Flux models.

A LoRA for faces. This is my first LoRA for a expression, so I'm still learning best practices for both training and prompting

Weights:

• I've found success with weights as low as 0.4, while other scenes have required up to 0.9. Mess around with it and find what works best for you.

Usage:

• I have found that a low weight in the prompt and a heavier weight in adetailer can work well. This can sometimes require specifying "eyes closed" or "eyes open" in both to prevent adetailer from trying to close an open eye (and vice versa), leading to weird results.

Issues:

• Tends to add eyeshadow. I trained another version to fix this but the issue persists. If it's a big enough problem, let me know in the comments.

If you have enjoyed any of my LoRAs and want to see more, consider supporting me. Anything helps!