Instagram: <a target="_blank" rel="ugc" href="https://www.instagram.com/synth.studio.models/">https://www.instagram.com/synth.studio.models/</a>Buy me a☕ <a target="_blank" rel="ugc" href="https://ko-fi.com/lonecatone">https://ko-fi.com/lonecatone</a>This represents many hours of work that I give away for free. If you enjoy it, Please make sure to 👍like and post to my models and workflows. That is how you support me!! Also, feel free to ⚡tip 😉<hr /><h3 id="v2.0">V2.0</h3>Man, I love this model. It is one of my pride and joys, but it had the worst face lock ever.This model goes back to my early Illustrious days, when I was just starting out. I baked a LoRA called <a target="_blank" rel="ugc" href="https://civarchive.com/models/915918/moriimee-gothic-niji-or-lora-style?modelVersionId=1244133">Moriimee Gothic Niji</a> into it. Man, it made sexy women (but it was really a single woman). It has been locked ever since.What changed:<ul><li>I trained in 224 images of different faces and ethnicities</li><li>I ran them through my face lock recognition software (in my ModelBuilder node pack) and removed any that hit over 50%</li><li>Wasted a ton of Buzz training on the old set by accident 🙄</li><li>3 trainings, 7 epochs tested. One winner: epoch 10, baked in at 0.8</li><li>Face lock went from 10.7% to 0.9% (how often two different people came out with the same face)</li></ul>Enjoy, and let me know what you think.<h3 id="v1.0">V1.0</h3><h3 id="i-wanted-color!!!">I wanted COLOR!!!</h3>I went on a qwest to create a checkpoint that reflects my Anime clashes with realism clashes with a "beautiful woman in a bikini weilding a Katana gets hit by a paint truck on the freeway" style. I focused on 🔎 details and color. Baked in a colorful modified VAE, and tweaked details and lighting.<h3 id="best-settings-(illustrious)">Best settings (Illustrious)</h3>Dpmpp++ Karras: Pulls out the colorEuler &amp; Simple: pulls out a lot of detail.Seeds3 &amp; Simple: Pulls off a ZIT look quite well 🤔Deis &amp; Simple: thumbs up 👍Deis &amp; sgm_uniform: oh wow 😍<h3 id="best-settings-(anima)">Best settings (Anima)</h3>Seeds3 &amp; Bong_tangent: winnerEuler &amp; Simple: pulls out a lot of detail.Seeds3 &amp; Simple: Pulls off a ZIT look quite well 🤔

Animosity | 二次元

V2.0Man, I love this model. It is one of my pride and joys, but it had the worst face lock ever.This model goes back to my early Illustrious days, when I was just starting out. I baked a LoRA called Morimee Goth into it. Man, it made sexy women (but it was really a single woman). It has been locked ever since.What changed:<ul><li>I trained in 224 images of different faces and ethnicities</li><li>I ran them through my face lock recognition software (in my ModelBuilder node pack) and removed any that hit over 50%</li><li>Wasted a ton of Buzz training on the old set by accident 🙄</li><li>3 trainings, 7 epochs tested. One winner: epoch 10, baked in at 0.8</li><li>Face lock went from 10.7% to 0.9% (how often two different people came out with the same face)</li></ul>Enjoy, and let me know what you think.

Files with SHA256: 08018fbf85931ce39b0d0f799eb1ef5f2154557c78eff06774bb684b8b7bd11c