Premium character LoRAs
Original synthetic characters, bundles and AI tools are in my shop:
Join the community on Discord: https://discord.gg/nB79wRNtMT
ELIAS CHATZOUDIZ INSP STYLE B BY RkRRa and STX
get the A version of ELIAS CHATZOUDIZ INSPIRED STYLE
with STX
Glossy Pin-Up Comic
A style LoRA for Krea 2 that turns any prompt into a glossy, saturated pin-up comic illustration: clean bold line art, shiny airbrushed skin, big expressive eyes, curvy figures and bright flat or patterned backgrounds. Great for bikinis, lingerie, costumes (cowgirls, nurses, mermaids, fairies, pirates) and NSFW pin-ups.
Usage
Start with the trigger word stxchatz, then describe the woman (hair, eyes), what she wears or that she is topless or nude, her pose and what she holds, and end with the background (a solid color, splatters, a beach, a desert...). Short, concrete captions work best: the trigger brings the style.
stxchatz, <woman> <clothes or nude>. <pose and props>. <background>.
Recommended weight: 1. Made for Krea 2 Turbo (8 steps, CFG 1).
Prompt Examples
"stxchatz, Woman with long brown hair wearing pink sunglasses, black and pink lingerie with a bow, purple thigh-high stockings, and black high heels. She stands with one hand on her hip and the other touching her face. White background with pink splatter effect."
"stxchatz, Blonde woman with a bob haircut and green eyes, wearing a black lace corset, black thong and white thigh-high stockings. She holds a martini glass, one hand on her hip. Pink background with white speckles."
"stxchatz, Mermaid with long blonde hair and a green and gold fish tail, topless, sitting on a rock by the sea. She holds a seashell, smiling. Blue water and a lighthouse in the background."
Training
Base model: Krea 2
Type: Style 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.