Adhésion à PixAI
PixAI
PixAI
Generate
Sign in
Accueil
Génération
Modèles
Studio
NEW
Concours
Créer avec Mio.2
Boîte à outils
NEW
Document PixAI

SDXL

AutismMix SDXL

Créer

  • Générateur d’images IA
  • Générateur d'animation IA
  • Boîte à outils
  • Générateurs thématiques
  • Entraîner un LoRA
  • Agent Mio.2
  • Studio

Découvrir

  • Tags tendance
  • Classement
  • Marché des modèles
  • Concours
  • Nouvelles

À propos

  • Document PixAI
  • Comment utiliser PixAI
  • Tsubaki.2
  • Découvrir Mio
  • Règles du contenu

Tarifs et aide

  • Adhésion
  • Packs de crédits
  • Contact

Application mobile

  • Télécharger l'application PixAI
  • App Store
  • Google Play
© 2026 PixAI
  • Conditions d'utilisation
  • Politique de confidentialité
  • Politique de droits d'auteur
  • Attributions
  • Loi sur les transactions commerciales spécifiées
AutismMix SDXL - AI Model cover image
Formé par

user-1722485577728538405

Téléchargé à
18 févr. 2024, 06:16
Utilisations
3.18m
Avis
Excellent  (275)
Étapes d'échantillonnage

25

Méthode d'échantillonnage

Euler a

Échelle CFG

7

Permissions
  • Autoriser la génération et le partage d'images
  • Permettre aux utilisateurs de télécharger votre modèle
  • Utilisations commerciales

LoRAs recommandées

Jinu style - AI Model cover image

LoRA utilisateur

Jinu style
674
Redline for PonyXL - AI Model cover image

LoRA utilisateur

Redline for PonyXL
19
Bigger Girls PonyXL - SyMix - BBW LyCORIS - AI Model cover image

LoRA utilisateur

Bigger Girls PonyXL - SyMix - BBW LyCORIS
361
ExpressiveH (Hentai LoRa Style) エロアニメ - AI Model cover image

LoRA utilisateur

ExpressiveH (Hentai LoRa Style) エロアニメ
674

Description

https://civitai.com/models/288584

The most impressive model I have ever had the pleasure of using. Can understand danbooru and e621 tags but can also understand natural language. Capable of anime, furry and realistic stuff. The model knows thousands of characters and people, even obscure ones.

Always use these tags at the start of a prompt: score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up

Use these tags to change style: source_anime, source_cartoon, source_furry, source_pony

AutismMix_pony here: https://pixai.art/model/1715773341038047866

(Description of LoRA on civitai page):

Mix of pony with some stuff. It's an attempt at making pony more predictable and less dependent on schizo negatives without removing its comprehension and artist knowledge. Personally I'm using AutismMix_confetti for general use and AutismMix_pony for certain loras. If you want to train a lora on top of autism I recommend doing so in the AutismMix_pony version for better compatibility.

I was asked if I will be making more versions and the answer is yes, but I want every version to be compatible with most of the PonyV6 loras and styles while keeping the large array of fetishes base pony can do. It's very easy to dumb down the model when its merged too much as you can see from other pony mixes, and I don't want to create another AOM situation.

What is the difference between the models:

AutismMix_confetti is a small amount of AnimeConfettiTune and AutismMix_pony. Has less style swing than pony and better hands. I prefer this one.

AutismMix_pony is a merge of ponyv6 with loras, its more compatible with certain styles made for the base ponydiffusion model.

AutismMix_DPO is AutismMix_confetti+DPO lora, made by request. Very similar to confetti version.

Add 3d to negs if you want a more traditional anime style. Quality tags should be same as ponyv6, but feel free to experiment: "score_9, score_8_up, score_7_up, score_6_up, score_5_up, score_4_up, source_anime, BREAK"

From my testing schizo negatives and those negative embeds made for SDXL/pony make it worse, but do whatever you want.

Commentaires6