A LoRA that is meant to help generate<h1 id="heading-1018">The Japanese Adult Video (JAV) Actress Hitomi Tanaka</h1>Girl has a lot of weight on her shoulders, go support her with follow, like, etc. <a target="_blank" rel="ugc" href="https://linktr.ee/hitomi_official">Here</a>.LoRA trained on the <a target="_blank" rel="ugc" href="https://civarchive.com/models/2661?modelVersionId=15640">URPM</a> model, because I found the results to be quite good and close to the training data.<h1 id="heading-6">Useful information:</h1>As usual, use Hires. fix It is pretty much mandatory (at least 1.5 times, but 2 times is what I recommend when it comes to 512x768, or 512x1024). Download free VRAM here: <a target="_blank" rel="ugc" href="https://th.bing.com/th/id/OIP.CykaFcF-5V34WbsJfnoOVgHaEK?rs=1&amp;pid=ImgDetMain">click</a>As mentioned, <a target="_blank" rel="ugc" href="https://github.com/Bing-su/adetailer">ADetailer</a> to fine tune the face and eyes. The eldritch hands/fingers we just gonna have to live with 😔.For the other options such as CFG, denoising strength, etc. refer to what is recommended by the model you're using. Or experiment and see what works for your individual case.<h1 id="heading-1434">Disclaimer:</h1><h3 id="heading-1435">This Lora is quite volatile. Fair warning, it can produce both, and in equal measure, amazing results along hilarious messes that make you question reality.</h3>I suspect it's due to Hitomi's massive assets. (<s>or my incompetence as a LoRA/model trainer</s>) They are so voluminous that the model is confused where they begin, where they end and how to fit all of them in a single frame. That, sometimes leads to rather unique body shapes. Also rather odd number of assets, too.Anyways, don't take it too seriously, try it, and if the result is terrible, just generate again. It's fun, kind of like lottery, you know?Cheers!

PulenKompot-Hitomi Tanaka

This is version 2, yes. Version 1 doesn't exist. It got deleted by an unknown omniscient cosmic force/entity.Current version (v2) is trained on URPM-v1.3

Files with SHA256: 4e3700e9fc796bef8c00398c35c03d83eebbbdec129efe665c4bb52a4ca98176