User LoRA

XL
Estilo

Zenith | Style v0.1 | Ilv1.0

Zenith | Style v0.1 | Ilv1.0 - AI Model cover image
Entrenado por
sc

Shane Creds

Entrena tu propio LoRA
Subido en
31 jul 2025, 11:52
Utiliza
156
Ganado
40,557
Reseñas
Excelente  (1)
Palabras clave

painterly, textured, dramatic_strokes, ze—nith12

Permisos
  • Allow image generation & sharing
  • Permita que los usuarios descarguen su modelo
  • Usos comerciales

Descripción

⚠️ Experimental ⚠️

Experimenting the optimal originals to be used for final dataset that has 100 images. This used 32 originals times 3 for augments, then I added 4 originals to match 100.

Recommended: 0.5–1.5 weight Any Illustrious models Euler/A/DDIM/DPM++ (any) any cfg will work if your model handles them well !!DON'T REMOVE THE ACTIVATION TAGS!!

If you want hires — use any reso then use 0.9 weight and 12–35 steps.

Comentarios