PixAI Üyeliği
PixAI
PixAI
Generate
Sign in
Ana Sayfa
Oluşturma
Modeller
Recipe
NEW
Studio
Yarışma
Mio.2

Conversational creation

Araç Kutusu
Guide

User LoRA

SD

Disco Elysium Character

#Azur Lane#Disco Elysium
Ana SayfaModellerMio.2Profil

Yapay Zekâ Araçları

  • 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

Keşfet

  • Trend Etiketler
  • Sıralama
  • Model Pazarı
  • Yarışma
  • Haberler

Hakkında

  • PixAI Hakkında
  • Guide
  • PixAI Nasıl Kullanılır
  • Tsubaki.2
  • Mio ile tanışın
  • İçerik Kuralları

Fiyatlandırma ve Yardım

  • Üyelik
  • Kredi Paketleri
  • İletişim

Mobil Uygulama

  • PixAI uygulamasını edinin
  • App Store
  • Google Play
© 2026 PixAI
  • Hizmet Şartları
  • Gizlilik Politikası
  • Telif Hakkı Politikası
  • Attributions
  • Belirtilen Ticari İşlemler Kanunu
Disco Elysium Character - AI Model cover image
Yüklenme Tarihi
2 Eyl 2023 09:19
Kullanım
15.6k
Yorumlar
İyi  (6)
Tetikleyici Kelimeler

zaum, oil painting, portrait

İzinler
  • Allow image generation & sharing
  • Kullanıcıların modelinizi indirmesine izin verin
  • Ticari kullanımlar

Recommended LoRAs

Disco Elysium / Aleksander Rostov /  Оil paints & Abstraction / Style - AI Model cover image

User LoRA

Disco Elysium / Aleksander Rostov / Оil paints & Abstraction / Style
26
FURRY BABES [Pony & illustrious LoRAs] - AI Model cover image

User LoRA

FURRY BABES [Pony & illustrious LoRAs]
10
Ryouske Hayashi  - AI Model cover image

User LoRA

Ryouske Hayashi
0
Winx Club Style lora - AI Model cover image

User LoRA

Winx Club Style lora
10

Açıklama

LoRA capable of generating character portraits in the style of Disco Elysium, as close to the originals as possible.

Trained with EveryDream2Trainer on the Lyriel v1.6 model, on 85 Disco Elysium character portraits (100 epochs), each one is out-painted and manually edited to a 512×512 resolution.

Lyriel v1.6+ model is recommended, but it should be compatible with other models.

Use LoRA weight of 2 for more abstract/impressionistic character portraits (think “Egg Head”), and 1 for something less surreal.

Notes:

Some prompts easily “overweight” the LoRA, so use weight >1 if it seems that it has no effect

Keep in mind that default negative prompts destroy LoRA’s intentionally imperfect style. That includes textually-inverted negatives, like “easynegative”.

All image examples were generated with HiRes-fix, Foolhardy Remacri upscaler, with a 0.4 denoising strength.

Yorumlar