This is a LoRA trained on my own AI-generated images, compatible with the MOSS int8-base format.
Please use this LoRA with the following settings:
・Model: moss_image_base_int8_convrot
・Trigger word: MK480U
・LoRA strength: 1.0
・Skip layer guidance: 26
・Sampler: ER_SDE
・Scheduler: SGM_Uniform
In addition to the trigger word, I recommend using prompts written in natural English sentences.
Additionally, due to the nature of the training data, images of men tend to have shorter proportions (i.e., they appear shorter) than they would in reality.
If necessary, try adding terms such as "tall man" or "adult man."



















