상세 정보
추천
v4 XL
v3
v2
v1
Real ????? - Butterfly

Real ????? - Butterfly

44.6K
327
1.1K
#여성
#개념
#vulva
#??????

Join our Discord for more images and upcoming updates!


Real ????? - Butterfly v4 XL


This is a rework of v3 with more images and trained on SDXL. I have included bigger ????? images this time for better variance. The LoRA now consists of 415 images.

Model 1: Big Lust v1.0 (SDXL) Brighter images (I have used this for training)

Model 2: Big Lust v1.5 (SDXL) Darker images with minor quality loss

SDXL ≠ Pony You should use SDXL models for generation if you want realistic results.

Training Resolution

  • Resolution 1: 832x1248, 2:3, Portrait

  • Resolution 2: 1024x1024, 1:1, Square

  • Resolution 3: 1248x832, 1:1, Landscape

Training time: 16.5h (RTX 4090)

For tags see v3


Real ????? - Butterfly v3


V3 was trained on 340 images from butterfly ????? in three different zoom levels. All images are hand picked, cropped and improved manually. I have removed some older images and added new images to the set. The training setting have changed and the tagging is a bit more simple now. They follow my other ????? LoRA so it's easier to mix them.

This picture shows what the LoRA was trained of. So don't expect a butterfly ????? in doggy position or full body standing settings. Use the main keywords and tags to get those results.

Scale 1 is used to generate the base image. Scale 3 can then be used to inpaint the ????? to get better details or fix minor issues. Scale 2 fills the gap for medium zoom levels.

You can see most images are trained for Scale 3 so inpainting is my major use of this LoRA.

Training Resolutions

  • 627x941 (2:3) (Portrait)

  • 768x768 (1:1) (Square)

  • 941x627 (3:2) (Landscape)

627x941 is probably unusual but because of an issue in Kohya this is the only way I can get portrait and landscape images trained in the same bucket. I prefer to use 640 & 960 as next values for generation and all my images using this values work fine.

Model: Lazymix Real Amateur 3.0a

Note: For testing purpose I have also trained LoRAs on Lazymix v4.0 but found the results with v3.0a slightly better. You can generate with v4.0 but since the merge is different compared to v3.0a you will get different results.

Inpainting

If you generate images that have lower resolutions as the training size then the results will look bad. Same applies for the inpainting target and area. If that is too small the result will look weird. It can help to upscale the image before you inpaint the pussy.

Important: Use an inpaint version of Lazymix Real Amateur 3.0a.

Follow this guide.

Preview images

The preview images are upscaled with hiresfix (2.25) to get to1440 x 2160. A few images have minor inpainting to fix details.

Tags

Keep in mind: Tags raise the chance to get what you want. But depending on the amount of training images containing this tags some tags might need more luck than others to work.

Main tag

pussy_scale1, pussy_scale2, pussy_scale3 - See description above

Zoom level

close up, extreme close up - Fits the scale levels, scale 1 has no zoom tag

Pussy

showing ????? - No hands and fingers on the ????? lips

If you want that use my Spreading LoRA

Labia size

????? having [size] inner labia

size: small, medium, long

Seems still to be very random.

Shaving

shaved vulva or (line, triangle, ...) + shaped pubic hair

... with stubble - Add this suffix or place in negative

Arousal

dry ?????, wet pussy

Other common tags

butterfly ????? - All images, can be used like a keyword

woman - All images

photo - Most images

fingers - If they are visible in the image

butthole - If visible in the image

rendering - A few images were made by SD

Example prompts

pussy_scale1, photo, woman, brown hair, small ??????, showing ?????, front view, butterfly ?????, ????? having small inner ?????, dry ?????, shaved vulva, ????????, lying on a towel, outdoor location, blurry background

pussy_scale1, photo, woman, brown hair, small ??????, showing ?????, front view, butterfly ?????, ????? having medium inner ?????, dry ?????, wide line shaped pubic hair, ????????, lying in bed, brown wall in background

pussy_scale2, close up, photo, woman, showing ?????, front view, butterfly ?????, ????? having medium inner ?????, dry ?????, shaved vulva, ????????, sitting on a chair with a teal pillow

pussy_scale2, close up, photo, woman, showing ?????, front view, butterfly ?????, ????? having medium inner ?????, dry ?????, shaved vulva, fingers, wearing yellow top, sitting in gras in garden

pussy_scale3, extreme close up, photo, woman, showing ?????, front view, butterfly ?????, ????? having medium inner ?????, dry ?????, shaved vulva, ????????

pussy_scale3, extreme close up, photo, woman, showing ?????, front view, butterfly ?????, ????? having long inner ?????, wet ?????, line shaped pubic hair, wearing clothes, lying on fur blanket


Real ????? - Butterfly v2


V2 was trained on 275 images from butterfly ????? in close up and also a few body shots. All images are hand picked, cropped and improved manually. I have removed the 'small' from the name as this version also has normal and longer ????? included.

Training Resolution

  • 627x941 (2:3) 51% (Portrait)

  • 768x768 (1:1) 25% (Square)

  • 941x627 (3:2) 24% (Landscape)

Model: Lazymix Real Amateur 3.0a

Preview images are generated with 2.25x HighRes fix with minor inpainting

How to use this LoRA?

Follow the link to see tagging details and how to mix this LoRA with my other LoRAs.

Example prompts

pussy_butterfly, extreme close up photo of a woman showing her ?????, front view, small butterfly ?????, ????? having short outer ????? and short inner ?????, dry ?????, shaved vulva with stubble, ????????

pussy_butterfly, extreme close up photo of a woman showing her ?????, front view, normal butterfly ?????, ????? having medium outer ????? and short inner ?????, moist ?????, shaved vulva, ????????, sitting on couch

pussy_butterfly, photo of a woman with brown hair and small ?????? showing her ?????, front view, normal butterfly ?????, ????? having medium outer ????? and medium inner ?????, dry ?????, wide line shaped pubic hair, ????????, lying in bed, brown wall in background

pussy_butterfly, close up photo of a woman showing her ?????, front view, small butterfly ?????, ????? having medium outer ????? and short inner ?????, dry ?????, shaved vulva, white finger nails, lying on blanket

pussy_butterfly,  photo of a woman with brown hair and medium ?????? showing her ?????, front view, big butterfly ?????, ????? having short outer ????? and medium inner ?????, moist ?????, triangle shaped pubic hair, ????????, pink finger nails, necklace, lying in a chair

pussy_butterfly, extreme close up photo of a woman showing her ?????, front view, big butterfly ?????, ????? having medium outer ????? and long inner ?????, moist ?????, shaved vulva, ????????, sitting on a yellow surface

Have a look at the promo images for more examples.


Real ????? - Butterfly v1


Introduction

This is my third release after long testing with LoRA generation. I hope you like her.

I am planing to release other LoRA with similar tags in the future so you can mix different shapes and create new styles.

Training Data

This LoRA was trained with close up images of similar ????? to get a consistent look.

Image count: 67

Resolution: 512x768

I will probably update her in the future as I got more images I wan't to include.

Model

LazyMix+ (Real Amateur ?????) v2

I might train this LoRA on other models too to see if the results will be better. For now it seems that only this model can be used for generation so don't waste your time with others.

For inpainting use the same model but converted it to an inpaint version with the checkpoint merger of webui. Inpainting this LoRA without an inpainting model will be hard.

Don't forget to switch back to the regular model if you get weird results. Happens to me all the time when I start a new image. The inpaining model can not produce normal images.

Tags

pussy_butterfly_small - Main tag

butterfly ????? - ????? is spread like a butterfly

butthole - Changes poses so buttholes are better visible

photo - Training image was a real photo

rendering - Training image was generated with an earlier LoRA version (experimental)

How to use

DPM++SDE Karras: More realistic images

Euler a: Smoother images

Use 512x768 for image size and upscale later.

CFG scale depends on the tags you are using. Anywhere in the range from 2 to 6 should work. Lower values are better.

Try to avoid Restore Faces if possible as it will sometimes changes the ????? to some kind of weird face or adds teeth.

For basic close up images just add the LoRA + pussy_butterfly_small tag. For more tags see the LoRA metadata. If you add too much tags the result may get worse.

For body shots you need to lower the LoRA strenght but this will reduce ????? details. I try to use the Regional Prompter extension so I can keep the strenght higher. Fix anything with inpainting that you don't like. Or just create an image without the LoRA and inpaint the ????? with LoRA afterwards.

Also works good to inpaint on real HD images to replace a ?????. This was the primary purpose why I created this LoRA.

The close up sample images are created by txt2img without editing. The full body images are completely inpainted from real images.

Upscaling

For best results I am using ControlNet with type tile + Ultimate SD upscale to keep the image stable without adding ghost ????? all over the place. Upscale 2-3x works most of the time.

번역문 보기

평점 및 리뷰

4.3 /5
0 개의 평점

충분한 평가나 댓글을 받지 못했습니다.

L
Lucifie
30
1.8K
공고
2023-06-19
모델 게시
2025-01-27
모델 정보 업데이트
모델 상세정보
유형
LORA
게시 날짜
2025-01-27
기본 모델
SDXL 1.0
트리거 단어
See tags description
복사
버전 소개

Trained on XL with more images and new settings.

허가 범위
모델 출처: civitai

1. 재게시된 모델의 권리는 원 제작자에게 있습니다.

2. 모델 원작자가 모델을 인증받으려면 공식 채널을 통해 SeaArt.AI 직원에게 문의하세요. 저희는 모든 창작자의 권리를 보호하기 위해 노력합니다. 인증하러 이동

창작 허가 범위
온라인 생방송
혼합 진행
다운로드 허용
상업적 허가 범위
생성된 이미지를 판매하거나 상업적 목적으로 사용 가능
모델의 재판매 또는 융합 후 판매 허용
QR Code
SeaArt 앱 다운로드
모바일에서 AI 창작 여정을 계속하세요