SECOND VERSION FOR FLUX.1 dev

Can do soft NSFW pretty well on its own: ???? modeling/posing, ??????? (mostly), female genitalia (somewhat)

Much better prompt adherence, much more true to the non-Flux model style. Can do well with both anime and realism, as well as other styles. Applies more typical CicaStyle features more often (shiny skin, see-through, anatomical proportions, etc.). Trained with a keyword but seems to work well enough without it.

• I run between CFG 1-3.5, Euler, 15-30 Steps

• I tend to use prompts closer to Pony/SD comma style vs Flux verbose, but I think a mix of both tends to work best on average, e.g. SDXL prompting style: "A woman (doing x), wearing a bodysuit, perfect face and eyes", etc.

• I deliberately tagged with long descriptive phrases
  

Overview:

• Good all-around Lora made from a lot of different styles and types (3d, anime, cartoon).

• GREAT FOR SHINY SKIN/OUTFITS AND OUTFIT DETAIL

• Can do bold lines well, in addition to soft edges/gradients like those seen in 2.5d/3d images. Almost like a Reiq-style for 3d, Jojo's for cartoon/anime, flat for anime, but also flexible.

• Seems to work well and mix with other Style Loras

Fishnets, bodysuits, and skin tight clothing come recommended.