사용자 LoRA

XLDiT.2
디테일 강화

Anime Hands- Improved

#들고 있기

#쥔 주먹

#편 손

Anime Hands- Improved - AI Model cover image
업로드 날짜
2026. 3. 24. 오전 5:06
이용수
16.97k
획득
200,000
리뷰
훌륭함  (1)
트리거 단어

anime_hands, hands, vroid-style, high_detail, v-hands, v-sign, holding, 2h_holding, holding_weapon, holding_gun, open_hands, carded_hands, fists, pointing, pointed_finger, finger_horns, hand_sign, peace_sign,

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

설명

Improvement over my earlier hand model lora test- which was a splendid success.


trained on 99 images (1 was corrupted so one of the staff angles is missing) poses trained on:

  1. neutral hands (t-pose, open, thumb down)
  2. clenched fist
  3. 1 hand holding staff/rod (ends of staff out of shot)
  4. 2 hands holding sword/ katana
  5. pointed finger/pointing index finger
  6. flipping the bird (middle finger)
  7. V-sign aka peace sign
  8. rocker hands aka devil horns
  9. holding a gun (finger on trigger)
  10. 'carded' hand (all fingers together, straight out)

Recommended settings: Base Model: trained on Noob AI XL (any SD XL should work), -Steps: 16-50, -CFG: 4.0 to 12.0, -Lora Weight: 0.4 to 1.0 (more towards 1 but not 1.0 for best effect), -workspace handfixer can help or hurt, depends on image enhance output,

댓글