If you like my work, drop a 5 review and hit the heart icon. it's free and keeps me motivated<h2 id="all-my-models-are-officially-hosted-and-maintained-by-me-on-tensor.art-.-use-my-exclusive-and-public-model-for-free-on-tensor.art-eeq913jiq">All my models are officially hosted and maintained by me on <a target="_blank" rel="ugc" href="http://Tensor.art">Tensor.art</a> . use my Exclusive and public model for free on <a target="_blank" rel="ugc" href="http://tensor.art">tensor.art</a></h2><h3 id="v5.0-civitai-onsite-generation-enabled-:-040125-t75gquk9a">v5.0 civitai onsite generation enabled : 04/01/25</h3><h3 id="v5.0-final-release-:-010125-6vfl76nb1">v5.0 final release : 01/01/25</h3>v5.0 pre relelase update : 25/12/24Check <a target="_blank" rel="ugc" href="http://tensor.art">tensor.art</a> for V1,V2,V3 online generation support<h3 id="you-can-run-prefect-pony-xl-and-use-its-api-on-sinkin-7t3mg6rbn">You can run Prefect Pony XL and use its API on <a rel="ugc" href="https://sinkin.ai/m/6zv9aDj">SinkIn</a></h3>currently, only v5 and v4 are available for civitai generation. v3,v2,v1 are retired from civitai generation. check other version details for alternative on site generation service. WIP: Realistic<h3 id="v5-update-:-3dk13unaf">V5 Update :</h3>onsite generation : <a target="_blank" rel="ugc" href="https://tensor.art/models/815661199413897287">https://tensor.art/models/815661199413897287</a>pre-release : <a target="_blank" rel="ugc" href="https://tensor.art/models/810892802167877454">https://tensor.art/models/810892802167877454</a><ul><li>Added trained Lora</li><li>Cleaned style</li><li>Baked in VAE</li><li>Extra trained data of 15000 steps</li><li>Better Lora support.</li></ul><h3 id="v4-update-:-zo4vtl05o">V4 Update :</h3><ul><li>onsite generation : <a target="_blank" rel="ugc" href="https://tensor.art/models/794513755405953653/Prefect-Pony-XL-v4">https://tensor.art/models/794513755405953653/Prefect-Pony-XL-v4</a></li><li>Added trained Lora</li><li>Cleaned style</li><li>Baked in VAE</li><li>Extra trained data of 15000 steps</li><li>Better Lora support.</li></ul><h3 id="v3-update-:-in42v0bc5"> V3 Update :</h3><ul><li>onsite generation <a target="_blank" rel="ugc" href="https://tensor.art/models/772139385270444718">https://tensor.art/models/772139385270444718</a></li><li>added some style to it</li><li>should give better face and eyes</li><li>better NSFW supports</li><li>wide range of Lora support</li></ul><h3 id="v1-lightning-:-ep418cndn">V1 lightning :</h3><ul><li>8-steps</li><li>DPM++ 2M</li></ul><h3 id="v2-update:-zk16vgh71">V2 update:</h3><ul><li>onsite generation <a target="_blank" rel="ugc" href="https://tensor.art/models/754887012743423435">https://tensor.art/models/754887012743423435</a></li><li>has less style influence</li><li>can work with wide range of loras</li><li>better anatomy</li><li>better color</li></ul><h3 id="v1-update:-46il4dpa5">V1 update:</h3><ul><li>On site generation closed for v1 on civitai for limitation of civitai.</li><li>try <a target="_blank" rel="ugc" href="http://tensor.art">tensor.art</a> if you want to use the online generation for v1</li><li><a target="_blank" rel="ugc" href="https://tensor.art/models/724940954567461228/Prefect-Pony-XL-v1.0">https://tensor.art/models/724940954567461228/Prefect-Pony-XL-v1.0</a></li></ul>If you like my work then drop a 5 review and hit the heart icon. it's free and keeps me motivated<h3 id="-wk7b1pwct"></h3>Suggested settings:<ul><li>I had CLIP skip 2 on every image</li><li>I had ENSD: 31337 all of them</li><li>All of them had highres.fix or img2img at higher resolution.</li><li>I don't use restore faces</li><li>I use afterdetailer for face details</li><li>Tiled diffusion for img2img upscaling and sometimes Noise Inversion for some more extra details.</li><li>4x-Ultrasharp upscaler</li></ul>NOTE: if you find any prompt of the preview images look familiar it's because I've taken them from other model review images, Credits to the original authors. Thanks for the benchmark.<h3 id="commission-open-in-patreon.-tczr9qqzt">Commission open in <a target="_blank" rel="ugc" href="https://www.patreon.com/hinokiart">Patreon</a><a target="_blank" rel="ugc" href="https://www.patreon.com/goofyai">.</a></h3><h3 id="get-early-access-and-exclusive-nsfw-lora-in-my-patreon-.-ec4b3afdy">Get early access and Exclusive NSFW Lora in my <a target="_blank" rel="ugc" href="https://www.patreon.com/hinokiart">Patreon</a><a target="_blank" rel="ugc" href="https://www.patreon.com/GoofyAi"> .</a></h3>Support my work by joining any one of them and get early access to all my upcoming loras and other perks such as fan requests and Discord role.<h3 id="join-my-discord-server-y8zexxmdn">Join my <a target="_blank" rel="ugc" href="https://discord.gg/M8yAsU9ZhC">Discord Server</a></h3>

Prefect Pony XL

<ul><li>Lighting improvement </li><li>Better style</li><li>Closer to v2 and v3 style composition </li></ul>

prefectPonyXL_v50.safetensors

CassRedmv2

Cassredmv1

v5.0

v4.0

v1-lightning

v2 cleaned style

v1.0

Suggested settings:<ul><li>I had CLIP skip 2 on every image</li><li>I had ENSD: 31337 all of them</li><li>All of them had highres.fix or img2img at higher resolution.</li><li>I don't use restore faces</li><li>I use afterdetailer for face details</li><li>Tiled diffusion for img2img upscaling and sometimes Noise Inversion for some more extra details.</li><li>4x-Ultrasharp upscaler</li></ul>

Files with SHA256: d1a2aac8e5b4ae14397620f29f8d3d953a7cb30d0a0e658781ded1120c6b8f3c