V2 update: this version of lora was trained with text encoder and should have even more prompt comprehension and be more close to the artist style. All training tags can be found <a target="_blank" rel="ugc" href="https://files.catbox.moe/7rifq2.txt">here</a>This one is best to use with some shibari/bondage/ball-gag/blindfold/pantyhose prompting. Recommended using it with autismmixes with short negatives.I recommend to use it with <a target="_blank" rel="ugc" href="https://github.com/Haoming02/sd-webui-vectorscope-cc">vectorscope extension</a> with setting from example images if your gens becomes too bright or unsaturated.

Himitsu style (for pony XL)

Old version trained without TEnc.Some of the examples images was generated with <a target="_blank" rel="ugc" href="https://github.com/Zuellni/stable-diffusion-webui-image-filters">image filters extension</a> with color 2, contrast 1 settings, if you see your gens becomes too white, I can recommend to use it too

Files with SHA256: 94bb683ccb3ad1dd08c56017c055ccf74f2361b845c627693342741a08058413