<h1 id="underview-v1-a-penis-model-for-z-image">Underview v1 - A Penis Model for Z-Image</h1><p>This model is version 4 of attempting to train male anatomy with z-image turbo. I used the z-image de-turbo model provided by Ostris's Ai-Toolkit. After many attempts, I found that training at a high step count (this model is at 12k steps with 50 images at 512px with the de-turbo preset) and essentially over-baking allowed a low strength to be used so that quality of the image isn't badly degraded. This model is best used with a strength of 0.50-0.67, the quality should be similar to base (or better) and remains very flexible in outputs. It works best with a CFG more than 1 but less 3, I usually use 1.8 at 12-20 steps. You may need to do a low de-noise after to get some finer details. The model was trained with captions provided with the <a target="_blank" rel="ugc" href="https://github.com/hydropix/AutoDescribe-Images">AutoDescribe-Images</a> tool provided by hydropix on github. The trigger word is Underview, I specifically avoided training images of the top part of the glans as loras tend to get confused on which way is top or below, so keep vantage points at a lower angle and expect to get gens with the penis pointed up or to the side.<br /></p><p></p><h1 id="underview-v2">Underview v2</h1><h3 id="v2-is-a-modified-version-of-v1-with-an-improved-and-much-more-consistent-output.-thanks-to-the-amazing-realtimelora-v2-node-shootthesound-made!"><br />v2 is a modified version of v1 with an improved and much more consistent output. thanks to the amazing <a target="_blank" rel="ugc" href="https://github.com/shootthesound/comfyUI-Realtime-Lora">RealTimeLora v2</a> node shootthesound made!</h3><p>After trial and error, the layers 24, 27, 28, and 29 could be turned off and eliminate most of the unwanted noise/artifacts that are present when the lora gets to full strength.</p><p>The lora also seems to understand higher angle views now, keeping it named "Underview" may be underselling it.</p><p>The RealTimeLora node pack also comes with a lora scheduling node that greatly improves the output.</p><p>The best schedule for the lora is (0:0, 0.2:0.65, 0.85:1, 1:0.65).</p><p>The v2 workflow is here: <a target="_blank" rel="ugc" href="https://civarchive.com/models/2218381/underview-z-image-penis-model-workflow">Underview v2 Workflow</a></p><p>The v1 version of the lora is still recommended to be used as a base pass (0.65 strength) before passing it to the v2 lora (with the above strength schedule).</p><p>v1 is better at following the prompted body types if you are wanting a less defined physique.</p><p></p>

Underview - Z-Image Penis Model

<p>v4 for me but v1 for release, focused on male anatomy from a view looking up</p>

Files with SHA256: 75b19a95e5af9ad60e2d3dab1ca7c9c21017ae11e514f234f006b30715975d0e