<p><span style="background-color:rgb(23, 23, 23);color:rgb(183, 183, 183);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">this model is good at creating that ideal Brunette that you dream of having as your next door neighbor.  this model is really good at body formation and face also but has trouble when creating a pose where she has to look back at the camera sometimes, with her backside showing. not all are failures, but works better for certain aspects than others depending on the scheduler. haven't found that absolute sweet spot yet, but karras or Beta using DPM++_2M_SDE seems to work best with 30-40 steps and 4.0-7.0 cfg. the lora is trained on the Lustify APEX v8 checkpoint and was trained with a completely fictional data set from created from one of my other character Loras created here on CivitAi but never published it.  </span></p>

Files with SHA256: 6a14fdc8567e09f28d48872456e98a8902b814109cf8f2e9f45f3bf94c295519