This model trained with less sample. Lowest the LoRA scale for better experience
Take 0.6-0.8
Gerald__Alchemist_Code.safetensors