PixAI Membership
PixAI
PixAI
Generate
Sign in
Home
Generation
Models
Studio
NEW
Contest
Create with Mio.2
AI Tools
NEW
Guide

SDXL

Mergestein - 3D

#3DCG

Create

  • AI Image Generator
  • AI Animation Generator
  • AI Tools
  • Thematic generators
  • Train LoRA
  • Mio.2 Agent
  • Studio

Discover

  • Trending Tags
  • Ranking
  • Model Market
  • Contest
  • News

About

  • Guide
  • PixAI Blog
  • Tsubaki.2
  • Meet Mio, Our Mascot
  • Content Rules

Pricing & Help

  • Membership
  • Buy Credits
  • Contact

Mobile App

  • Get the PixAI app
  • App Store
  • Google Play
© 2026 PixAI
  • Terms of Service
  • Privacy Policy
  • Copyright Policy
  • Attributions
  • Specified Commercial Transactions Law
Mergestein - 3D - AI Model cover image
Uploaded at
Dec 28, 2025, 6:33 AM
Uses
37.6k
Reviews
Excellent  (6)
Sampling Steps

13

Sampling Method

Euler

CFG Scale

2

Negative

worst quality, bad quality, low quality, lowres, anatomical nonsense, artistic error, bad anatomy, interlocked fingers, bad feet, poorly drawn, censored, watermark, bad hands, extra digit, fewer digits, sketch, jpeg artifacts, bar censor, mosaic censor, conjoined

Permissions
  • Allow image generation & sharing
  • Allow users to download this model
  • Allow commercial use of generated images

Recommended LoRAs

Diana (Pragmata) - AI Model cover image

User LoRA

Diana (Pragmata)
124
Cute face style - AI Model cover image

User LoRA

Cute face style
1.74k
Hyper SDXL - AI Model cover image

User LoRA

Hyper SDXL
8.57k
MGC1_1993998480931592836 - AI Model cover image

User LoRA

MGC1_1993998480931592836
0

Description

available for download at CivitAi: https://civitai.com/models/2256150?modelVersionId=2539698

Recommended to add best quality, masterpiece on positive prompt (if you want: realistic and photorealistic prompt works too)

Good quality at low steps.

I needed to re-merge the model, so it works better here in pixai (I reduced details and realism a little), the first version looked bad, this new one looks better but I recommend using HyperSDXL LoRa to use low steps (less than 13 steps).

Comments1