v1.0 Release!
LoRA was trained on, and all images were generated with, Counterfeit v2.5
All examples use the seed of "1" with a batch size of 8, meaning no seed cherry picking was done.
v1.0
Trained on 160 character portraits from Royal, Strikers and Vanilla versions of Persona 5. The dataset was hand picked to have 1 of each character expression and outfit combination, so this release should have a wider variety of those. LoRA strength of somewhere around 0.6 - 0.8 seems to work the best.
CFG: 8
Sampler: DPM++ SDE Karras
Resolution: 512x512
Go ahead and upload yours!
Your query returned no results β please try removing some filters or trying a different term.