Premium character LoRAs
Original synthetic characters, bundles and AI tools are in my shop:
Join the community on Discord: https://discord.gg/nB79wRNtMT
Latex Shiny
A concept LoRA for Krea 2 focused on women in shiny latex, PVC and vinyl, NSFW: catsuits, corsets, bodysuits, opera gloves and thigh-high boots, deep gloss and bright reflections, cutouts and open zippers. It gives an amateur photo look.
Usage
Start with the trigger word LatexShinyRK, then describe the garment first and precisely (catsuit, hood, corset, bodysuit, dress, leggings, opera gloves, thigh-high or platform boots, buckles and zippers), its colour, how glossy it is and the reflections on it, how tight it fits, what is exposed in plain words through cutouts or open zippers, her hair, makeup and expression, the pose, the place, and end with the light. Naming the gloss and the reflections brings out the effect the most.
LatexShinyRK, Photograph of a woman in a <glossy black latex catsuit with an open zipper down to her navel>, <bright reflections along her hips>, <kneeling on a bed>, in a <bedroom>. The lighting is <warm lamp light>.
Recommended weight: 1. Made for Krea 2 Turbo (8 steps, CFG 1).
Prompt Examples
"LatexShinyRK, Photograph of an adult woman with light skin and long blonde hair, seated on a black chair in a wooden room. She is wearing a black, glossy, tight latex bodysuit with a high-cut design, exposing her buttocks and upper thighs."
"LatexShinyRK, Photograph of an adult woman with fair skin and long black hair with red highlights, styled straight and loose. She wears a red, glossy latex nurse outfit with a tight corset that exposes her breasts, small to medium in size, and nipples."
"LatexShinyRK, Photograph of an adult woman with a light tan skin tone and long, wavy blonde hair. She is wearing a black, glossy latex corset with garter straps and thigh-high black stockings."
Training
- Base model: Krea 2
- Type: Concept LoRA
- Steps: 4000
- Rank: 32
- Optimizer: adamw, constant learning rate
- Engine: ai-toolkit
Generation settings
Trained on Krea 2, and works on Krea 2 Turbo for faster generation.
- Raw: ~30 steps, guidance 3.5
- Turbo: ~8 steps, guidance 1.0
The gallery images were generated with krea2_turbo_mxfp8.safetensors, at 8 steps, guidance 1, LoRA weight 1.



















