Sampler：DPM++2M Karras     steps： 38     512X1024   cfg：7Hires. fix： X2，SwinIR 4x step：25 Denoising：0.26hope u like it! more my model  <a rel="ugc" href="https://www.liblib.art/userpage/1e3c4c0fc33545f89056e8b46523d3da/publish">https://www.liblib.art/userpage/1e3c4c0fc33545f89056e8b46523d3da/publish</a> more my pics <a target="_blank" rel="ugc" href="https://www.liblib.art/userpage/1e3c4c0fc33545f89056e8b46523d3da/publish/image">https://www.liblib.art/userpage/1e3c4c0fc33545f89056e8b46523d3da/publish/image</a>If you are unable to replicate my example image, the reason is that the random seed of the sd-webui where I generated the image is generated from the CPU, and the default setting is GPU. You can try changing it.

繁花|写真-13

13_v10.safetensors

v1.0

Flowers | Photographs-13

&nbsp; &nbsp; &nbsp; Try it online. I've released a few non-public test versions before, they ran quite well locally, but tend to crash online, especially with the recent testing layers, which caused one solid-state drive to be written off. Luckily, there are partial backups of some models on the other solid-state drives. I'm just not very interested anymore.&nbsp; &nbsp; &nbsp; Feel free to try it out. The clarity of Miss 13 might not be the highest, but I quite like the overall colors, many of which don't need post-editing in Photoshop. Some scenes come with a slight film effect. Feel free to explore this based on your interests.

Files with SHA256: 789167480c079401042a198af022727d5a2c253c7c86efae107eeba41f1b4bd7