Model Information:Base model: IllustriosXL 2.0Sampling Type: EpsilonSampling Method:<ol><li>Euler A</li><li>Euler Dy (Recomended)</li><li>Euler DY CFG++</li></ol>Schedule Type:<ul><li>Auto</li><li>Normal</li><li>Exponential</li><li>React Cosinusodinal Dynsf</li></ul>Resolution<ul><li>1440x2560 (2K)</li><li>1080x1920</li><li>Other default Stable Diffusion Resolution:<ul><li>1024 x 1024 (1:1 Square)</li><li>1152 x 896 (9:7) &amp; 896 x 1152 (7:9)</li><li>1216 x 832 (19:13) &amp; 832 x 1216 (13:19)</li><li>1344 x 768 (7:4 Horizontal) &amp; 768 x 1344 (4:7 Vertical)</li><li>1536 x 640 (12:5 Horizontal) &amp; 640 x 1536 (5:12 Vertical)</li></ul></li></ul>Future Work:<ul><li>Change base model to NetaYume Lumina 3.5</li></ul>Q &amp; AQ: Why stop training on Illustrious?A: Limitation model.Q: Can generate NSFW? A: You're Committed a sin, if you REALLY wanna do it, just do it.Q: Is there v-prediction?A: NOQ: Newest dataset? A: no, ++knowledge -&gt; hard to maintain resolutionChangelog:0.3 (Same as 0.3-TEST before)<ul><li>This model will be deprecated. Thanks for using my model. don't be sad, there better than our model &gt;:)</li></ul>0.3-TEST0.220.210.20.150.1<h3 id="license:">License:</h3><a target="_blank" rel="ugc" href="https://huggingface.co/Laxhar/noobai-XL-1.0/blob/main/README.md#model-license">NoobAI</a> <a target="_blank" rel="ugc" href="https://freedevproject.org/faipl-1.0-sd/">IllustriousXL</a><span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);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">If you'd like to support my work, you can gimme coffee<a target="_blank" rel="ugc" href="https://paypal.me/IRedDragonICY">https://paypal.me/IRedDragonICY</a><a target="_blank" rel="ugc" href="https://ko-fi.com/ireddragonicy">https://ko-fi.com/ireddragonicy</a><a target="_blank" rel="ugc" href="https://saweria.co/IRedDragonICY">https://saweria.co/IRedDragonICY</a><span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);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">(Actually, I don't need money, just need happiness)

DragoXL

a major revision of the DragoXL paradigm. This version mostly addresses fundamental visual integrity problems, especially malformed objects and scene inconsistencies, so producing more coherent, accurate, and aesthetically pleasing outputs.Future research will keep emphasizing on improving fine-grained control, so strengthening prompt understanding for abstract ideas, and so extending the model's capacity in challenging compositions and narrative coherence.

Files with SHA256: e113eceab2dd6640e95affae293f542cb63be591ae89afc332e746d78a2635be