PixAI 멤버십
PixAI
PixAI
Generate
Sign in
홈
생성
모델
Studio
NEW
콘테스트
Mio.2 창작
툴박스
NEW
가이드

사용자 LoRA

XLDiT.2
복장

Vacuum-Sealed Clothing

#수영복#레오타드#바디수트

제작

  • AI 이미지 생성기
  • AI 애니메이션 생성기
  • 툴박스
  • 테마 생성기
  • LoRA 학습
  • Mio.2 에이전트
  • Studio

찾아보기

  • 인기 태그
  • 랭킹
  • 모델 마켓
  • 콘테스트
  • 뉴스

소개

  • 가이드
  • PixAI 블로그
  • Tsubaki.2
  • Mio 소개
  • 콘텐츠 규칙

요금 및 도움말

  • 멤버십
  • 크레딧 팩
  • 연락처

모바일 앱

  • 앱 스토어
  • 구글 플레이
© 2026 PixAI
  • 개인정보
  • 개인정보 처리방침
  • 저작권 정책
  • 특정상업거래법
Vacuum-Sealed Clothing - AI Model cover image
제작자
jw

John West

나만의 LoRA 학습하기
업로드 날짜
2026. 3. 26. 오후 9:34
이용수
4.6k
획득
260,016
리뷰
훌륭함  (2)
트리거 단어

v4cuumcl34v4g3, skintight, separated breasts

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

추천 LoRA

UI test (DiT.2) - AI Model cover image

사용자 LoRA

UI test (DiT.2)
0
Yukimura Kusunoki (楠 幸村) - Haganai: I don't have many friends (僕は友達が少ない) - AI Model cover image

사용자 LoRA

Yukimura Kusunoki (楠 幸村) - Haganai: I don't have many friends (僕は友達が少ない)
12
Ingrid - Swedish woman. Huge breasts. SFW or NSFW - AI Model cover image

사용자 LoRA

Ingrid - Swedish woman. Huge breasts. SFW or NSFW
6
Cow 牛 - AI Model cover image

사용자 LoRA

Cow 牛
13

설명

Want to show the curves without showing the skin? This "impossible clothing" LORA tucks underneath and hugs around and between the girls, pulls in the navel and booty and clings to muscles.

Trained on Illustrious and should work with Illustrious-based models. Well with Haruka. Will be testing other models.

The girls are usually a bit perky, hence the NSFW tag. Using the negative tags "covered nipples" and "cameltoe" can attempt SFW but the lora will ignore them randomly, especially at higher values.

Works as intended at most positive strengths, with strengths higher than 1.0 sometimes needed for clothes/LORA it struggles with.

Strength seems to affect faces and bodies without prompts. Higher strengths create faces and bodies in the 20s, younger at lower values, and straight-up cartoonish at high negatives. Leans towards large breasts at higher strengths and smaller at lower. Some models may override these tendencies.

This LORA will usually prioritize prompts and other LORAs over its own tendencies at strength of 1.3 and lower. Very high strength (typically 1.7 and up) causes it to consistently ignores prompts that alter height and body age (but uses older and younger faces, it's weird).

"Muscular," "chubby," "slender," and such all function without issue if the model understands the prompt (i.e. most models don't understand "slender hips").

Works with many snug clothes. Inconsistent with zippers, buttons and flaps (it likes to open them). Didn't find any issues with colors or materials (provided the model understands).

No issues yet with view and camera angles.

Still testing combining with other character, clothing and artstyle LORAs, but initial tests seem to have few conflicts.

댓글