사용자 LoRA

XL
스타일

stitchheart ノワール

#중성적인

#주얼리

#사이버 패션

stitchheart ノワール - AI Model cover image
업로드 날짜
2025. 9. 15. AM 3:46
이용수
1.50k
획득
200,000
리뷰
굿  (3)
트리거 단어

goth, dewy base, big glassy eyes, aegyosal, gradient under-eye blush, jelly gloss lips, soft cel + painterly mix, subtle halation, soft rim light

권한
  • 이미지 생성 및 공유 허용
  • 다른 사용자가 내 모델을 다운로드하도록 허용합니다
  • 생성된 이미지의 상업적 이용 허용

설명

A glossy-eyed emo/grunge portrait style that blends sharp ink lines with soft painterly skin. Stitchheart pushes porcelain micro-texture, dewy base, smoky lids, jelly-gloss lips, and high micro-contrast under cinematic lighting. Think manipulative siren energy—chains, safety pins, cross jewelry, ribbon lacing, and low-rise silhouettes.

What it does best

  • Semi-realistic anime faces with huge glassy eyes (aegyosal highlight + gradient under-eye blush)
  • Soft dark eyeshadow + obsidian tightliner; clustered lower lashes
  • Crisp lineart with subtle film halation/rim light
  • Grunge / goth / Y2K / mcbling accessories (chains, pins, ribbons, hardware)
  • Works for girls (long, ultra-straight hair + side bangs) and guys (short mullet/wolf cuts, teasing gaze)

Grunge hardware & body detail (optional toggles)

safety pins embedded into arm, chain hardware, grommet belt, tiny cross pendant layers,
dermal ribbon-lacing on lower waist, ribbon-laced dermal anchors on lower waist, hip bone ribbon lacing,
surface anchors with satin ribbon, small bow tied to one side, midriff exposed, low-rise waistband

Recommended Negatives

child, teen, loli, young-looking, school uniform, flat 2D shading, lowres, bad anatomy, extra fingers,
overexposure, heavy HDR, watermark, text, signature

Prompt tips

  • Girls: “1girl, long straight hair, side bangs, black eyes, glossy lips, soft dark eyeshadow”
  • Guys: “1boy, man, masculine body, short mullet haircut, teasing smirk, manipulative gaze, lifted hem, low-rise jeans”

Good settings

  • Aspect: portraits 4:5 / 1:1 / 9:16
  • Size: 768–1024 px on the short side
  • CFG: 5–7 | Steps: 28–36
  • LoRA/Model strength: 0.6–8.0

댓글