<h3 id="vpred-models-are-trained-with-v-parameterization-so-they-will-only-work-on-vpred-models-like-noob-ai-xl-vpred-or-obsession-etc.-i3oog850l">VPred models are trained with V-Parameterization so they will only work on VPred models like NOOB-AI-XL-vpred or Obsession, etc.</h3>This is a DoRA so make sure to have your stuff somewhat updated<h2 id="remember-to-check-out-the-artist!!!-without-their-art-these-loras-wouldn't-exist!!!-3qt62q2ro">Remember to <a target="_blank" rel="ugc" href="https://deumosden.newgrounds.com/">check out the artist</a>!!! Without their art, these LoRAs wouldn't exist!!!</h2>NoobAI and Obsession could replicate their style on their own, however, the style is clearer when using a separate model like this one.

DMXwoops – DeumosDMX (Artist Style) | NAI-XL/Illustrious

<ul><li>Initial version trained on 1240 images with Noob-AI-XL VPred1.0</li><li>I recommend using <a target="_blank" rel="ugc" href="https://civitai.com/models/820208?modelVersionId=1198119">Obsession</a></li><li>Generally works but requires Hires.fix for better results</li><li>I dunno, but something makes me think that I should also train these style DoRAs for EPS Illustrious models since a lot of people also use those (WAI-NSFW-illustrious, Hassaku XL, Noob-AI-XL Epsilon-pred 1.1, and so on). If you're actually reading this far, tell me in the comments if you actually want that</li></ul>

Files with SHA256: ba3a27dbbe86097a1956051d2043230dbef0c6fe8e25e25d962edb4f263f4305