This LoRA was trained over a 100% synthetic none copyrighted custom dataset.

Recommended settings

• Strength: 0.8

• Sampler: Res_multistep

• Scheduler: simple

• CFG: 1

• Steps: 8