---
tags:
- text-to-image
- stable-diffusion
library_name: diffusers
inference: false
---

# SDXL Evolved - Aesthetic

<div style="overflow:hidden">
  <img src="imgs/1.png" width=256 height=256 style="float:left; margin:0"/>
  <img src="imgs/2.png" width=256 height=256 style="float:left; margin:0"/>
  <img src="imgs/3.png" width=256 height=256 style="float:left; margin:0"/>
  <img src="imgs/4.png" width=256 height=256 style="float:left; margin:0"/>
</div>
<div style="overflow:hidden">
  <img src="imgs/5.png" width=256 height=256 style="float: left; margin:0"/>
  <img src="imgs/6.png" width=256 height=256 style="float: left; margin:0"/>
  <img src="imgs/7.png" width=256 height=256 style="float: left; margin:0"/>
  <img src="imgs/8.png" width=256 height=256 style="float: left; margin:0"/>
</div>
Images generated with no prompt.

## Generation settings

Best results with:
Euler with trailing(SGMUniform) at 8+, Guidance scale 1. Similar to SDXL Lightning

## Details

Created with [https://github.com/255BITS/diffusion-evolver](https://github.com/255BITS/diffusion-evolver)

Evolved for 20 cycles towards more aesthetic images.