

This LoRA is trained to mimic the art style of Besuyama. All sample images are straight out of ComfyUI with a Hires Fix pass, no inpainting or Adetailer was used. The dataset contains 303 images, which were manually de-watermarked, with a mixture of automatic and manual tagging.
Trigger Word: besustyle
Additional Tags: adult, bara, muscular
Weight Range: 1.0 recommended, but can be adjusted between 0.6 - 1.0 depending on the base model.
NOTE: This dataset contains images. If you do not want , make sure you actively prompt away from it. Also, some models may introduce unintended alterations, so ensure your prompts align with onsite generations when generating content. To avoid unwanted content, ensure to use child, in your negative prompt, and if you like 'em beefy, add bara in your positive prompt.
Please do not violate the CivitAI content policy -> https://civarchive.com/safety#content-policies
Initial rerelease since I'm dumb and deleted my old account with the panic.