Stable Diffusion Xl

DreamShaperXL1.0

model_id: dreamshaperxl10

<p><strong>Finetuned over SDXL1.0.</strong><br>Even if this is still an alpha version, I think it's already much better compared to the first alpha based on xl0.9.<br>For the workflows you need Math plugins for comfy (or to reimplement some parts manually).<br>Basically I do the first gen with DreamShaperXL, then I upscale to 2x and finally a do a img2img steo with either DreamShaperXL itself, or a 1.5 model that i find suited, such as DreamShaper7 or AbsoluteReality.</p> <p><strong>What does it do better than SDXL1.0?</strong></p> <ul> <li> <p>No need for refiner. Just do highres fix (upscale+i2i)</p> </li> <li> <p>Better looking people</p> </li> <li> <p>Less blurry edges</p> </li> <li> <p>75% better dragons 🐉</p> </li> <li> <p>Better NSFW</p> </li> </ul> <p>&nbsp;</p> <h3 id="heading-226">Old DreamShaper XL 0.9 Alpha Description</h3> <p>Finally got permission to share this. It's based on SDXL0.9, so it's just a training test. It definitely has room for improvement.</p> <p>Workflow for this one is a bit more complicated than usual, as it's using AbsoluteReality or DreamShaper7 as "refiner" (meaning I'm generating with DreamShaperXL and then doing "highres fix" with AR or DS7).<br><br>Results are quite nice for such an early stage.<br><br>I might disable the comment section as I'm sure some people will judge this even if it's early stage. I also don't think this is on par with SD1.5 DreamShaper yet, but it's useless to pour resources into this as SDXL1.0 is about to be released.</p>