SDXL

vaelsem

vaelsem - AI Model cover image
Subido en
29 dic 2025, 20:59
Utiliza
1.7k
Reseñas
Justo  (1)
Pasos de muestreo

30

Método de muestreo

Euler a

Escala CFG

3.3

Excluir de la imagen

wallpaper forced, bad perspective, 3d, petite, skinny, worst quality, low quality, lowres, what, jpeg artifacts, (bad), error, artistic errors, ai-generated, (loli, shota:1.9), bad anatomy, bad hands, fused fingers, wrong hands, bad feet, bad teeth, extra digits, watermark, grey hair, displeasing, unfinished, generic, early, draft, signature, text, logo, username, deformed legs, deformed limbs, extra legs, digits,(, title), patreon, greyscale, distortion, fisheye, distorted, warped, perspective distortion, see-through clothes, revealing clothes, old, oldest, misplaced genitals, multiple views, sharp teeth, shark, cross print, aged down, face markings, particles, distortion, text

Permisos
  • Permitir la generación y el intercambio de imágenes
  • Permitir que los usuarios descarguen su modelo
  • Usos comerciales

LoRAs recomendadas

Descripción

A couple of weeks after my first merge, I've had this itch on the back of my ear to give it a go again, and here I am. This mix has been a huge surprise to me, and it feels good, it gens good so far, we'll see from now on. And hopefully you guys enjoy it as well, I'd love some feedback.And to my surprise, this checkpoint is even good enough with natural language (although half the time and a couple of gens later), I usually can get pretty close to a almost complete scenario by just describing my booru prompt, which is good enough for me I guess, I was able to come up with some gens that would be hard to get by just prompting tags, so yeah...Quality tags I've been using: newset, amazing quality, very aesthetic, absurdres, eyecatch, newest, 2024, high detail, incredibly detailedRecommended settings I've tested so far:Euler | Euler A | RestartCFG 3 to around 4.5Around 28 to 36 steps (max I've went so far)If you want to support me, you can either donate, drop a like or follow me for future updates, either one helps. Thank you for all the support.LicensesSDXL - CreativeML Open RAIL++-MIllustrious - https://freedevproject.org/faipl-1.0-sd/NoobAI-XL - https://civitai.com/models/license/1140829Copied these licenses from Hyphoria's main page. Respectfully adhering to the same base licenses from these models.

Comentarios

Obra de arte hecha por vaelsem