<h3 id="absynth-enhanced-stable-diffusion-3.5-base-models-vr0wyxt23">Absynth - Enhanced Stable Diffusion 3.5 Base Models</h3>Absynth is a series of specialized variant Stable Diffusion Base Models, meticulously crafted through the fusion of custom trained Negative LoRAs, fine-tune-merged inversely to expand and enhance the model's ability to generate hyper-detailed images. Versions are available for Stable Diffusion 3.5 Large &amp; Stable Diffusion 3.5 Medium at various sizes on this page as well as a <a target="_blank" rel="ugc" href="https://civarchive.com/models/982393/absynth-sdxl">SDXL</a> version of Absynth that has its own page.NOTICE: 2.0 is the best current version.<h3 id="wip-okmsezo3g">WIP</h3>Absynth is a work in progress being fine-tuned on less-than optimal hardware, new versions will be shared as meaningful progress is made. We are further refining Absynth’s unique training approach to more detailed results.<h3 id="note:-ek5jf354o">Note:</h3>Absynth may sometimes produce over-detailed images or visual artifacts. Efforts are underway to improve stability and refine its output in future updates.<h3 id="credits:-az6d7l0x6"> Credits:</h3>Absynth was created by Doctor Diffusion in partnership with FreeUse AI.

Absynth - Enhanced Stable Diffusion 3.5 Base Models

I find the fp8 version to be a little less overcooked and works better but I will keep this here. Corrections to the <code>t5_max_length</code> for SD3.5L training resulted in this improved test version created in conjunction with FreeUse AI. Suggested settings:<ul><li>steps: <code>40</code></li><li>cfg: <code>4.5</code></li><li>sampler: <code>dpmpp_2m</code></li><li>scheduler: <code>sgm_uniform</code></li></ul>

Files with SHA256: 63c53bbc1793e660cdc133e61fbef4e90e31caa52da6283a2506889c0e49284c