PixAI-Mitgliedschaft
PixAI
PixAI
Generate
Sign in
Startseite
Generation
Modelle
Studio
NEW
Wettbewerb
Mit Mio.2 erstellen
Toolbox
NEW
PixAI-Dokument

Benutzer-LoRA

XL
Stil

ZONE Flash Style

Erstellen

  • KI-Bildgenerator
  • KI-Animationsgenerator
  • Toolbox
  • Themengeneratoren
  • LoRA trainieren
  • Mio.2 Agent
  • Studio

Entdecken

  • Angesagte Tags
  • Rangliste
  • Modell-Marktplatz
  • Wettbewerb
  • Nachricht

Über uns

  • PixAI-Dokument
  • So verwendest du PixAI
  • Tsubaki.2
  • Mio kennenlernen
  • Inhaltsregeln

Preise & Hilfe

  • Mitgliedschaft
  • Credit-Pakete
  • Kontakt

Mobile App

  • Hol dir die PixAI-App
  • App-Store
  • Google PlayStore
© 2026 PixAI
  • Nutzungsbedingungen
  • Datenschutzbestimmungen
  • Urheberrechtsrichtlinie
  • Gesetz über bestimmte kommerzielle Transaktionen
ZONE Flash Style - AI Model cover image
Hochgeladen am
13.04.2026, 19:46
Verwendet
341
Bewertungen
Keine Bewertungen
Auslösende Worte

z0n3_thick

Berechtigungen
  • Bildgenerierung und -freigabe erlauben
  • Nutzern erlauben, dieses Modell herunterzuladen
  • Gewerbliche Nutzung

Empfohlene LoRAs

Sandy | Mukatsuku Imouto wa Chanto Shikaranakucha | Juna Juna Juice artist | Doujinshi version | Illustrious/NoobAI - AI Model cover image

Benutzer-LoRA

Sandy | Mukatsuku Imouto wa Chanto Shikaranakucha | Juna Juna Juice artist | Doujinshi version | Illustrious/NoobAI
3
[ILXL] Kim Kwang Hyun Style - AI Model cover image

Benutzer-LoRA

[ILXL] Kim Kwang Hyun Style
4
Full Metal Panic! Characters and Style (Illustrious) - AI Model cover image

Benutzer-LoRA

Full Metal Panic! Characters and Style (Illustrious)
3
Kuroinu Art Style - IL - AI Model cover image

Benutzer-LoRA

Kuroinu Art Style - IL
9

Beschreibung

Read Me - New info on V2 !!!Trained on very high quality exports of some of Zone's Flash games:Hentaikeygirl 5Natsume 2Umeko 2Reiko 2Zone Tan from Natusme 2 and Umeko 2 openingThat one Rukia flash gameCleaned them up with FFDec (removed background and UI) and exported the most important (but not too many) scenes with 4000x4000 resolution and then Downscaled to (and trained on) 1024x1024.add watermark as negative - or (watermark:-1) in positive if you use negpipUse 0.8 strength. 1 can work too but some details might become scuffed such as nipplesI prefer E15 for close-up and upper body and E10 and E6 for full body (E2 works there too but it's too weak in my opinion). E15 struggles with detail in full bodyV2 Thick Trigger: z0n3_thickOptional: instead of z0n3_thick, use z0n3 for slightly thinner outlines and softer detailV2 Soft Trigger: z0n3Optional: instead of z0n3, use z0n3_thick for slightly thicker outlines but slightly worse detailsThick = resembles the older flash games with thick outlinesSoft = resembles the newer flash games with soft outlines (is more refined and has better detail)V2 is a lot stronger than V1 but both can be used wellPlease post the stuff you create down below, I'm curious to see what you guys came up with :3I'm already working on a separate Lora which mainly focues on Chracter and Scenes of those games. I already have a couple versions, but those are really scuffed. I'm trying to make one where all those characters and most scenes work well in one single Lora while also being able to just swap the characters around or even use completely different characters outside of the lora. Same with scenes. The versions I have right now are alright at this at most, but not good enough..

Kommentare