Can get decent results with simple prompts such as :<pre><code>a house by william eggleston, sunrays, beautiful</code></pre><pre><code>closeup portrait of a woman in a kitchen by william eggleston, beautiful</code></pre><pre><code>a beautiful view through a kitchen window, car, by william eggleston, sunlight</code></pre>Trained using <a target="_blank" rel="ugc" href="https://github.com/TheLastBen/fast-stable-diffusion">https://github.com/TheLastBen/fast-stable-diffusion</a> SDXL trainer.ComfyUI seems to give better results than A1111, but that's just me.

SDXL William Eggleston Style

<ul><li>Clearer and more coherent output</li><li>More accurate representation of the Eggleston style</li><li>Better aspect ration handling</li></ul>

Files with SHA256: c4196a84d15f3d5a7d660a989fd91a9ab0971c3a636f258c6a62dec52366b170