詳細
おすすめ
Futawarp Gooey Mix
Gooey Mix / Monstercock option

Gooey Mix / Monstercock option

521
196
1.5K
#NSFW
#ふたなり
#ディックガール
#NSFW
#ふたなり
#ディックガール

edit: added smaller safetensor versions, also realized I uploaded the wrong model for monster. Now changed the example prompts to match the first image for each respective model.

Merge: (((test furry_5 + .2 yiffy18) + .5 yohan diffusion) + .3 (some mixture of anyv3/airotic/futacum)) all using "weighted difference"

Merge of several different models that has decent capability for futa and furry content. Some of the models used have been forgotten.

The ones I do know are in here are: test furry5, yohan diffusion, yiffy18, anythingv3 and airotic pen.

because of all of the models merged together you'll frequently get images that have things floating or anatomically impossible features. Otherwise it's pretty capable with some good prompting.

one version is the base merge while the other is a 5050 weighted sum merge with monstercock. The monstercock merge is easier to create futa with as well as to do larger in general, the difference being that I believe it may sacrifice some ability to create anything other than a portrait of a hung lady. You're welcome to try merging the base with monstercock in different degrees. I've tried a .1 and .3 which both added to the quality of and increased the overall size of them. The base merge can do fairly good as well but I recommend including "vascular veins, frenulum, " as it seems to help making realistic looking members.

All of the images down below were done with heun sampler and most are latent upscaled a little with high res fix. 12.5 cfg.

Furry content is largely based off of Test furry5's trained data with yiffy to a much smaller degree. using "furry" will generally produce the strongest results but it struggles making true furry content as it'll tend to skip from cat-ears with some fur and a tail straight to just a cat or a human body with a cat head. So in my experience its good at kemonomimi with not much more but I haven't fully tested that. Perhaps going through and describing individual parts such as "body covered in fur" or "stomach covered in fur" would work to some degree.

e621 and booru tags work.

I recommend using the bad_artist and bad_prompt embedding as well as your own negs.

Needs a Vae. I use the anythingv3 vae as this has anyv3 as a part of its merge.

Prompt from the first image for gooey mix:

woman with a () lifting the front of her short skirt, small (natural) above a toned stomach, visible within deep , narrow waist, wide , (symmetrical frenulum, vascular veins, virile shaft),
((modern masterpiece, best quality, highly detailed art, incredible shadows), sensual skin, realistic, hyper realism, photorealistic:1.4), (detailed green eyes, distinct pupils, neon and cyberpunk background, edgy:1.1),
cinematic lighting, realistic reflections, humored expression, stark eyes, jawline, high cheekbones, black and lace and punk clothes, skin tight fishnetting, punk jacket, thigh highs,

neg:

lowres, (((text))), error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, jpeg artifacts, signature, (((watermark))), out of frame, extra fingers, mutated hands, (poorly drawn hands), (poorly drawn face), (mutation), (deformed ), (ugly), blurry, (bad ), (bad proportions), (extra limbs), cloned face, (multiple views), greyscale, monochrome, three feet, four feet, extra feet, extra legs, extra arms, logo, (((patreon username))), (((artist name))), (((signature))), (censored), bad hands, disembodied, fat, cross eyed, (bad perspective:1.2), (bad drawn eyes), deformed iris, simple iris, no pupils, 2d, monochrome, (bad shading:1.1), (simple background:1.3), extra , missing limb, floating limbs, disconnected limbs, malformed hands, long neck, long body, disfigured, amputee, mutated,

Steps: 20, Sampler: Heun, CFG scale: 12.5, Seed: 1196696380, Size: 768x1024, Model hash: 6bbf2dd6, Denoising strength: 0.5, Clip skip: 2, Hires upscale: 1.35, Hires upscaler: Latent

Prompt for first image of monster version:

hyperfuta, girl showing her (massive :1.1) and , embarrassed expression, futanari, vascular veins, (frenulum:1.1), hyper shaft, blushing, shy, cinematic lighting, newhalf, (detailed beautiful eyes, distinct pupils:1.1), (small ), subsurface scattering, fit, , , narrow waist, , (dark interior, detailed extravagant bedroom interior, starlit balcony background:1.1), , on the bed, solo focus, 1girl, explosively (from :1.1), (covered in , spray:1.1), (mindless euphoria, aroused:1.1), beautiful neon purple hair, (ineffectual ), ( straps), (pulled aside), pointed , puffy , ,
(masterpiece, best quality, highly detailed art, gorgeous perfect face, beautiful shadows, absurd res:1.5), (realistic, hyper realism, sensual skin, photorealistic:1.4), sweat sheen, slightly happy squint,

neg:

bad-artist, bad-prompt, lowres, (((text))), error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, jpeg artifacts, signature, (((watermark))), out of frame, extra fingers, mutated hands, (poorly drawn hands), (poorly drawn face), (mutation), (deformed ), (ugly), blurry, (bad ), (bad proportions), (extra limbs), cloned face, (multiple views), greyscale, monochrome, three feet, four feet, extra feet, extra legs, extra arms, logo, (((patreon username))), (((artist name))), (((signature))), (censored), bad hands, disembodied, fat, cross eyed, (bad perspective:1.2), (bad drawn eyes), deformed iris, simple iris, no pupils, 2d, monochrome, (bad shading:1.1), (simple background:1.3), extra , missing limb, floating limbs, disconnected limbs, malformed hands, long neck, long body, disfigured, amputee, mutated, (:1.1), extra ,

Steps: 20, Sampler: Heun, CFG scale: 12.5, Seed: 3537799613, Size: 768x1024, Model hash: 3a4db744, Denoising strength: 0.5, Clip skip: 2, Hires upscale: 1.3, Hires upscaler: Latent

翻訳を表示

評価とレビュー

4.2 /5
0 件の評価

まだ十分な評価やレビューが集まっていません

no-data
データなし
モデルと会話する
お知らせ
2023-04-01
モデルを公開
2026-01-31
モデル情報を更新
モデル詳細
タイプ
Checkpoint
投稿日時
2023-03-20
基本モデル
SD 1.5
バージョン紹介

use "vascular veins" and "frenulum" for better . using sometimes helps as well.

許可範囲
モデルソース: civitai

1.転載モデルは学習・共有目的のみで使用し、著作権は原作者に帰属します

2.モデルの認証は公式チャンネルでご連絡ください。クリエイターの権利保護に努めています クリックして認証

創作許可範囲
オンライン画像生成
統合
ダウンロード
商用利用の許可範囲
生成された画像は販売または商業目的での使用
モデルの転売やモデル統合後の販売
QR Code
SeaArt Appをダウンロード
モバイルでAI創作の旅を続けよう