Trained on 1024 res images and double blocks to help with motion and work better with other loras.

This lora needs a bit of nudging in the prompt to get what it's supposed to do. the facial features are mostly there but adding things like 'short hair' helps a lot.