Someone asked for Version 3. So here's Version 3.Version 3:Mixed with a mild style lora ontop of V2, it's still an illus v0.1 merged checkpoint. The colors aren't as deep as a Noob checkpoint might be, but I did my best to not try and alter its style too much. What's the point of <a rel="ugc" href="https://civarchive.com/models/934492/madly-mix-ver-nightnoob">NightNoob </a>otherwise? Cheers. ===============================================Version 2:Madly Mix Ver 1.0 merged with <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-nsfw-illustrious-sdxl?modelVersionId=1025184">WaiNSFWLustrious v5</a>Rolled in at .25 weight.More malleable to styles in general, it has some functionality for pony, but it largely has a nice mellow style.You can try it on <a target="_blank" rel="ugc" href="https://tensor.art/models/794022368197519974">Tensor</a>Version 1:It's been a little over a year since I first submitted a mix to Civitai, after fiddling around with a lot of mixes during the first part of SDXL and PDXL, Flux came, stuck around kinda, and now we have IllustriousXL.I was content to let other people do their magic, but I kept getting results I didn't quite want, or weren't quite to my taste, so I decided to have another go at making a checkpoint merge. You can use this model at <a target="_blank" rel="ugc" href="https://tensor.art/models/792220732496087964/Madly-Mix-Vol-Illustrious-vol-Illustrious">Tensor Art </a> Components:<ul><li>@bloodsplash 's <a target="_blank" rel="ugc" href="https://civarchive.com/models/868661?modelVersionId=991657">SpatterXLv1.5</a></li><li>My own style lorer cooked specifically for this mix</li><li>and <a target="_blank" rel="ugc" href="https://civarchive.com/models/827184/wai-nsfw-illustrious-sdxl">waiNSFWLustrious v3</a></li></ul>Essentials: Positive Prompt:<pre><code>masterpiece, best quality</code></pre>Recommended Negative Prompt:<pre><code>lowres, worst quality, low quality, bad anatomy, bad hands, 4koma, comic, greyscale, censored, jpeg artifacts</code></pre>What to expect:The outputs should generally be clean 2D, or 2.5D anime-style without needing to use things like 3d, realistic, photorealistic in the negs, or anime style in the positive prompts. I haven't tested any 3D or realistic images intentionally, since that's not really my speed, but you're welcome to if that's your jam.Cheers,Madly

Files with SHA256: 5c236bc9c63397bf8bbaf04d9600e18b93e0ba87bfe4106d760f2983319266e5