Iscrizione PixAI
PixAI
PixAI
Generate
Sign in
Home
Generazione
Modelli
Recipe
NEW
Studio
Contest
Mio.2

Creazione via chat

AI Tools
Guide

User LoRA

DiT.3DiT.2DiT.1XL
Other

ネットミーム 見せられないよ!-SDXL/Dit3/Dit2/Dit1

#Meme
HomeModelliMio.2Profilo

Strumenti IA

  • OC Maker
  • AI Anime Generator
  • VTuber Maker
  • Sonic OC Maker
  • Hazbin Hotel OC Maker
  • Anime Character Creator
  • AI Figure Generator
  • Genshin OC Maker
  • All AI Tools

Scopri

  • Trending Tags
  • Classifica
  • Market dei modelli
  • Contest
  • Notizia

Informazioni

  • Informazioni su PixAI
  • Guide
  • Come utilizzare PixAI
  • Tsubaki.2
  • Scopri Mio
  • Regole sui contenuti

Prezzi e assistenza

  • Appartenenza
  • Pacchetti di crediti
  • Contatto

App mobile

  • Scarica l'app PixAI
  • App Store
  • Google Play
© 2026 PixAI
  • Termini di servizio
  • Informativa sulla privacy
  • Politica sul copyright
  • Attributions
  • Legge sulle transazioni commerciali specificate
ネットミーム 見せられないよ!-SDXL/Dit3/Dit2/Dit1 - AI Model cover image
Addestrato da
Avatar image of Benxzeot

Benxzeot

Allena il tuo LoRA
Caricato su
16 set 2026, 23:41
Utilizzi
510
Guadagnato
58,662
Reviews
No reviews
Parole d'ordine

can't show this (meme), >_<, chibi, open mouth, holding sign, full body, standing, legs apart, standing on one left leg, japanese text "見せられないよ!",

Permessi
  • Allow image generation & sharing
  • Permettete agli utenti di scaricare il vostro modello
  • Usi commerciali

Descrizione

既存資材からの焼き直し 各モデル対応版 Dit3対応

推奨モデル順(再現度)SDXL>Dit3>Dit2>>Dit1 推奨適用強度:構図のみなら1.0、文字再現なら1.2。但し文字再現度はSDXL版でもそこまで再現出来ない 文字の入れ具合(変更)はDit3が一番いいだろう

学習資材側で持ち看板部分をなるべく平行にするよう調整してあるので 文字は自分で入れた方が何かと確実だろう

A rework based on existing training data. Versions available for various models.

Recommended model order (based on fidelity): SDXL > DiT-3 > DiT-2 >> DiT-1. Recommended strength: 1.0 for composition only; 1.2 for text reproduction. However, note that even the SDXL version cannot achieve perfect text accuracy. DiT-3 likely offers the best results for inserting (or modifying) the text.

Since the training data was adjusted to keep the handheld sign as parallel as possible, it is probably more reliable to add the text yourself.

Commenti