A general-purpose merge of some of the best SDXL models currently available. In my experience, it compares favorably to its constituent models.Version 1 was created by performing the following sequence of merges:<ol><li>50/50 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/5273/wyvernmix-15-and-xl">WyvernMix XL v1.1</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/128607/nightvision-xl-photorealistic-or-portrait-or-photography-or-hyperreal-or-architecture-or-interior-design-no-refiner-needed">NightVision XL Alpha v0.6.5.0</a></li><li>70/30 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/120964/rundiffusion-xl">RunDiffusion XL beta</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/112902/dreamshaper-xl10">Dreamshaper XL1.0</a></li><li>75/25 mix of v1.0 Mix #1 and v1.0 Mix #2</li></ol>Then, Version 2 was created by performing the proceeding further merges:<ol><li>50/50 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/129922">FormulaXL v1.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/129666?modelVersionId=142162">Realities Edge XL v1.0</a></li><li>65/35 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/123307?modelVersionId=140130">SDVN7-RealArtXL beta2</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/131703?modelVersionId=144779">BriXL v1.0</a></li><li>65/35 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/125703">ProtoVision XL Beta v0.5.2.0</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/120033?modelVersionId=138008">ChimeraMix(XL) v1.2</a></li><li>60/40 mix of v2.0 Mix #2 and v2.0 Mix #3</li><li>75/25 mix of v2.0 Mix #1 and v2.0 Mix #4</li><li>50/50 mix of v2.0 Mix #5 and Edvard v1.0</li></ol>Version 3 was created as follows:<ol><li>60/40 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/128607/nightvision-xl-photorealistic-or-portrait-or-photography-or-hyperreal-or-architecture-or-interior-design-no-refiner-needed">NightVision XL Beta v0.7.0.2</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/5273/wyvernmix-15-and-xl">WyvernMix XL v1.1</a></li><li>75/25 mix of v3.0 Mix #1 and v1.0 Mix #2</li><li>50/50 mix of v3.0 Mix #2 and v2.0 Mix #5 </li><li>50/50 mix of <a target="_blank" rel="ugc" href="https://civarchive.com/models/133005?modelVersionId=152076">Juggernaut v2</a> and <a target="_blank" rel="ugc" href="https://civarchive.com/models/122822?modelVersionId=133832">Crystal Clear XL</a></li><li>60/40 mix of v3.0 Mix #3 and v3.0 Mix #4</li></ol>I personally prefer to use the DPM++ SDE Karras sampler, the Real-ESRGAN X4+ upscaler, a CFG in the 4-6 range, and 30-45 steps, but your preferences may differ. HighRes Fix seems helpful. A refiner does not. I've made no effort to improve SDXL's NSFW capabilities here, and I don't plan to.

Files with SHA256: c6878f2fe9c13bcc9c7c697c5b756c767c4ae5f28e03f7cb73fea0e4fc246a51