Please support the original creator. Follow them here -> Twitter - Pixiv

This model was not created to harm or otherwise inconvenience the original creator nor Civitai or its users. It can be removed upon request.

Model based on the works of Hitboxsamurai on Pixiv and Twitter. Their work uses a mixture of stable diffusion generation and photoshop editing. Consider giving them a follow if you enjoy this style.

Model trained on WAI-IL-V9. You may use different checkpoints with this lora.

As you might expect, using tiled upscaling is bound to mess up the structure of the image.

Recommended settings (external gen-comfyui & civit generator):

• lora weight: 0.8-1.0

• sampler: euler a/euler a cfg pp

• scheduler: normal

• steps: 30-35

• CFG: 4.5-6.0/1.0-2.0

• hires fix

AI to AI !