Model
Trends
.ai
Model
Trends
.ai
les meilleurs modèles d'IA open source
Exécuter en local
windmillmixPony_wmmDark.safetensors
6.46 GB · SafeTensor · fp16 · full
Les téléchargements sont servis par Civitai. Certains créateurs demandent d'abord une connexion à Civitai.
AD
<windmillponydark-xl>
À propos de ce modèle
STABLE DIFFUSION XL 1.0 · Modèles de base
Créé par
yoshikage1048
sur Civitai
by yoshikage1048 - A gloomy pony merge - guidance 7 and 30 steps with clipskip 2 and euler a recommended. Example: score_9, score_8_up, score_7_up, zombie girl, silver hair, wedding dress, wedding veil, face veil, manor house, window, moonlight, looking at viewer, night, dark room, upper body, tied hair, hair over eye, from side, with negative -- score_4, score_5, score_6,
Notes du créateur
__________________________________________________________________________________________________ The Dark Side Update is coming! I redid the WMM merge with the addition of a marvelous model! → This merge is not just an upgraded version of WMM. It is an attempt at a different approach! So the value of the original version of WMM has not been lost! Feature - The merge is aimed at a darker atmosphere. - More subdued contrast than the original version of WMM...Maybe... (unless my eyes are corrupted... ) - Slightly more accurate depiction of fingers Notice The possibility of semi-realistic images being generated seems to be somewhat higher than in the original version of WMM. We will tell you how to deal with this case. - Add " anime," tag to the positives. - Use only “score_9,” in the score tag. Try one or both. If it still doesn't improve... - Add “realistic,” to the negative. (This is a fairly powerful tag, but it reduces the accuracy of the hand description somewhat.) Miscellaneous Upscaler "R-ESRGAN 4x+ Anime6B" may be a good choice. "4x_NMKD-UltraYandere_300k” is great, but seems to increase contrast too much. (If you know of a good upscaler that does not destroy the feel of the original image, please let me know!) Enjoy! And check out ! ...Prefer a darker atmosphere than this? Check out the ! __________________________________________________________________________________________________ Below is a description of the original version. I am not a native speaker of English. I use automatic translation. was somewhat inaccurate in its depiction of the fingers.(Not a problem with the original model. It is a problem with my merge.) The problem has been considerably improved by the v2 update! Check it out! → WindMillMix took care to ensure that the depiction of the fingers was accurate. Of course, MMM has not lost its value! It still has the potential to produce great images! Lizardon1025 for bringing this to my attention, thanks! This merge is a fusion of “ " and "”. Both are really great authentic models! NSFW images can also be generated. If there are any problems, this will be removed. __________________________________________________________________________________________________ Setting Example: (This is the setting I use. You may change it to your favorite setting.) Clip skip: 2 Steps: 30 CFG scale: 7 Sampler: Euler a or DPM++ 2M or DPM++ 2M SDE Schedule type: Automatic or Karras (After many attempts, the combination of Euler a and Automatic seems to be the least disruptive. However, the combination of DPM++ 2M SDE or DPM++ 2M and Karras may produce more creative images. Of course, this is my personal opinion and may not be correct. And, you can configure the settings as you wish. Please let me know your results.) Width: 896 or 832 Height: 1152 or 1216 VAE: none (Or Use your favorite one!) Hires.fix: 4x_NMKD-UltraYandere_300k (Recommend) or Your favorite one Upscale by: 2 Hires steps: 15 Denoising strength: 0.2~0.3 Positive Prompt : “score_9, score_8_up, score_7_up,” is fine. (Not including the “score” tag seems to result in a simpler image... Perhaps... Please check the comparison images in the post.) Negative Prompt : “score_6, score_5, score_4,” is fine. (Or it may be fine without including the score tag. I tried it several times and it made no significant difference.) The “realistic,” tag may not be necessary this time.This is because it is unlikely that semi-realistic images will be generated. The “monochrome,” tag still works. The tones of the image will be more intense. If the text is displayed, please add “text, signature, username, artist name,”. This may improve the situation. If you see a mosaic, add “censored, mosaic censoring, bar censor,”. It will be improved. Criticism, advice, ways to improve, all welcome! Waiting for lots of posts and reviews! _______________________________________________________________________ MMMはやや手指の描写の正確性に欠けていました。(元のモデルの問題ではありません。私のマージの問題です。) WindMillMixでは手指の描写が正確になるように気をつけました。 もちろん、MMMに価値がなくなったわけではありません！まだ、素晴らしい画像を生成する可能性があります！ Lizardon1025さんのおかげで気づきました、ありがとう！ 今回のマージは"Orinara · Pony AnimeMix"と"Mala Anime Mix NSFW PonyXL"の融合です。どちらも本当にオーセンティックな素晴らしいモデルです! NSFW画像の生成も可能です。 何か問題があれば削除します。 Setting Example: (これは私が使用している設定です。あなたのお気に入りの設定に変えても構いません。) Clip skip: 2 Steps: 30 CFG scale: 7 Sampler: Euler a or DPM++ 2M or DPM++ 2M SDE Schedule type: Automatic or Karras (何度も試行しましたが、Euler aとAutomaticの組み合わせが最も破綻が少ないようです。 ただし、DPM++ 2M SDEやDPM++ 2MとKarrasを組み合わせた場合、より独創的な画像が生成される可能性があります。 もちろんこれは、私の個人的な見解なので、正しくない可能性があります。 そして、当然ですが、あなたの好きなように設定をすることができます。成果を私に教えて下さい。) Width: 896 or 832 Height: 1152 or 1216 VAE: none (または、あなたのお気に入りのものを使ってください!) Hires.fix: 4x_NMKD-UltraYandere_300k(オススメです) or Your favorite one Upscale by: 2 Hires steps: 15 Denoising strength: 0.2~0.3 Positive Prompt : "score_9, score_8_up, score_7_up, "で問題ありません。 (scoreタグを含めない場合、よりシンプルな画像になるようです...恐らく... 比較画像をポストしているので確認してください。) Negative Prompt : "score_6, score_5, score_4, "で問題ありません。 (もしくは、scoreタグを含めなくても問題ないかもしれません。 何度か試しましたが、大きな違いはありませんでした。) 今回は"realistic,"タグは不要かもしれません。セミリアルな画像が生成される可能性が低いからです。 "monochrome,"タグはまだ働きます。画像の色調がより濃くなります。 文章が表示される場合は"text, signature, username, artist name, "を追加すれば改善する場合があります。 モザイクが表示される場合は"censored, mosaic censoring, bar censor, "を追加すれば改善します。 たくさんのポストとレビューを待っています！ 批判、アドバイス、改善方法、すべて歓迎です!
355 téléchargements · 51 j'aime sur Civitai
SHA256
F2C7E9D3C3EBF26BB4117907BED0EC976009126942688C55E63C66B323B480D9
ID de modèle ModelTrends.ai
#14403
Signaler le contenu
Rendus d'exemple
Prompts et réglages partagés par les personnes qui ont réalisé ces rendus sur Civitai. Choisissez un rendu pour voir comment il a été fait.
Réglages
Sampler
Euler a
Étapes
20
Guidance
5
Seed
2417725205
Taille
720x1280
Prompt
absurdres,highres,very aesthetic,ray_tracing, photorealistic, 1girl, from below, foreshortening, white shirt, red bow-tied neckerhicef, belt, teal skirt, standing, happy, highway, lightpole, outdoors, zPDXL3 <lora:Everlasting_Summer_Vanilla_Backgrounds_by_GraffMetal:1>, (1girl, aqua_hair,aqua_eyes, long twintails) <lora:Everlasting_Summer_Miku_by_GraffMetal:0.8>
Prompt négatif
nsfw, anime, source_anime,deep_negative_pony EZNegPONYXL-neg NEGATIVE_HANDS
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
545761117
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, zombie girl, silver hair, wedding dress, wedding veil, face veil, manor house, window, moonlight, looking at viewer, night, dark room, upper body, tied hair, hair over eye, from side,
Prompt négatif
score_4, score_5, score_6,
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
524761345
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, 2girls, outdoors, dark night, against wall, neon light, neon sign, looking at viewer, from above, shadow, BREAK white shirt, black skirt, pointed ears, blue skin, dark blue hair, pink eyes, BREAK white shirt, black skirt, demon, white hair, red eyes, pale skin, two ponytails with black bows, short hair, black wings, long eyelashes,collar,
Prompt négatif
score_4, score_5, score_6,
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
2757208465
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, 1girl, solo, orange hair, parted bangs, sailor uniform, blue eyes, focus on eye, constricted pupils, looking at viewer, head tilt, shaded face, forest, dark night,
Prompt négatif
score_4, score_5, score_6,
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
2988463065
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, girl, vending machine, snow, heavy breath, Beanie, scarf, sweater, from side, outdoors, looking at viewer, drop shadow, dark night, dim light,
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
477965738
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, girl, rabbit mask, sweater, overalls, black hair, long hair, dark night, forest, blood stains, filthy, holding axe, grin, covered eyes,
Prompt négatif
score_4, score_5, score_6,
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
1440570340
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, girl, black hair, long hair, parted bangs, seductive smile, red dress, black feather Necklace, Crow, dark night, forest, red eyes, glowing eyes,
Prompt négatif
score_4, score_5, score_6,
Réglages
Sampler
Euler a
Étapes
30
Guidance
7
Seed
744584614
Taille
832x1216
Clip skip
2
Prompt
score_9, score_8_up, score_7_up, girl, stained glass, Silhouette,
Modèles d'IA pour générer image, vidéo et texte
windmillponydark xl est un modèle de base pour la famille Stable Diffusion XL 1.0, référencé sur ModelTrends.ai, un catalogue en lecture seule de modèles d'IA open source pour l'image, la vidéo et le texte. Comparez-le aux autres modèles Stable Diffusion XL 1.0, consultez son score de chaleur pour suivre sa tendance et ouvrez-le sur PirateDiffusion ou BitVector pour l'essayer.
Parcourir par famille de modèles
Voir tous les modèles