사용자 LoRA

SD

Angrboda Style Furry LoRA (Anthro)

#퍼리
Angrboda Style Furry LoRA (Anthro) - AI Model 커버 이미지
업로드 날짜
2024. 1. 13. 오전 12:44
이용수
628
리뷰
리뷰 없음
트리거 단어

firefox, fox

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

설명

https://civitai.com/models/88146/angrboda-style-furry-lora-anthro

Disclaimer and Request Always credit Angrboda when posting images generated with this LoRA.

Trained on 100% publicly available images taken from e621.

Please do not sell this LoRA, or images generated with this LoRA (additionally, do not sell access to this LoRA, or access to images generated with this LoRA, e.g. via Patreon or other paid/subscription services).

Please do not remove metadata from images generated with this LoRA.

Model Info Trained on: CoffeeNSFW (My fav. anime model)

Epoch: 10

Repeats: 10

Resolution: 576

Images: 193 from e621

I specifically filtered out images with the following tags:

comic

monochrome

text

panels

older than 01/01/2018

Steps: 19300

ClipSkip: 2

Recommendations Trigger Words: None needed

Weight: Around 0.8 yielded good results

Samplers:

Euler A showed surprisingly strong results (again)

Otherwise DPM++ 2M Karras was my go-to

Base Resolutions: Around 384x576

Upscalers:

4x_fatal_Anime_500000_G showed the best results

4x_fooldhardy_Remacri also did OK, but more washed out

Latent (nearest-exact) is also an OK option, but only at very high denoise

Extensions:

Use Dynamic Thresholding Extension to use high CFGs without burn-in

by firefox_

댓글

Angrboda Style Furry LoRA (Anthro)으로 만든 작품