If you want to use more my checkpoint online generation, please visit here.<a target="_blank" rel="ugc" href="https://tensor.art/u/762555264535746522">https://tensor.art/u/762555264535746522</a><h2 id="v11-lbytu3yto">V11</h2>Improvement in overall quality, with more coherent composition and background.better hands and legs.<h3 id="recommended-setting-same-with-v10-hyaefmex7">Recommended Setting same with v10</h3>_______________________________________________________________________________________________<h1 id="v10-0shtmylqk">v10</h1>In order to differentiate from the REAL_CN , I changed the default face shape and overall style. Since many parameters were adjusted this time, it might not be as stable as previous versions. I'll work on optimizing it in the next version.<h3 id="v10-recommended-setting-stycr0c5u">v10 Recommended Setting</h3><ul><li>CFG scale: 7</li><li>Step:30-50</li><li>Sampler: Euler a</li><li>ADetailer face_<a target="_blank" rel="ugc" href="http://yolov8s.pt">yolov8s.pt</a> use (ADetailer sampler: DPM++ 2M, ADetailer scheduler: Karras ADetailer denoising strength: 0.35 )</li></ul> Positive Prompt<pre><code>score_9, score_8_up, score_7_up,</code></pre>Negative Prompt<pre><code>score_6,score_5,score_4,, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry, </code></pre>__________________________________________________________________________________________________v9To avoid making my models too similar, I’ve updated this one with a style that leans more towards a smoother 2.5D look.v9 Recommended Setting<ul><li>CFG scale: 7</li><li>Sampler: Euler a</li><li>ADetailer face_yolov8n/<a target="_blank" rel="ugc" href="http://s.pt">s.pt</a> use more better</li></ul>Positive Prompt<pre><code>score_9, score_8_up, score_7_up,</code></pre>Negative Prompt<pre><code>score_6,score_5,score_4,, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry, </code></pre>___________________________________________________________________________Positive Prompt<pre><code>score_9, score_8_up, score_7_up,</code></pre>Negative Prompt<pre><code>score_6,score_5,score_4,, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry, </code></pre>V5-&gt;V6-&gt;V7 change point<ul><li>Background quality improvement</li><li>More realistic</li><li>skin details</li><li>clothes details</li></ul>V3 v4 v5 v6 V7 change Recommended Setting<ul><li>CFG scale: 7</li><li>Sampler: Euler a</li><li>ADetailer face_yolov8n/<a target="_blank" rel="ugc" href="http://s.pt">s.pt</a> use more better</li></ul>v1 v2Recommended Setting<ul><li>Steps: 30</li><li>CFG scale: 3.5</li><li>Sampler: Euler a or DPM++ 2M Karras</li></ul>

WAI-REALMIX

waiREALMIX_v100.safetensors

v10.0

v9.0

v8.0

v7.0

v6.0

v5.0

v4.0

v3.0

v2.0

v1.0

我是主要在 civitai的活动的模型作者，因为一个契机也来这边发模型了，主要是制作pony相关的内容。 关于此模型此模型是pony的大模型，Lora只推荐pony基础的，SDXL的也不是不能用效果不怎么好。介于模型的特殊性，请一定要遵守国家相关法律，您基于本模型生成的内容由您自行维护并对其独立判断后使用，您需对生成内容负责，本人对由此造成的任何损失不负责任。 基础使用方法<ul><li style="text-align: start;">步数步数: 30</li><li style="text-align: start;">CFG : 7</li><li style="text-align: start;">采样: Euler a</li><li style="text-align: start;">如果中远景有面部问题 使用ADetailer</li><li style="text-align: start;">直接出图 请大于1024x1024</li></ul>(ADetailer sampler: DPM++ 2M, ADetailer scheduler: Karras ADetailer denoising strength: 0.35 )PS.打提示词的时候，请先不要一上来就给提示词加权重，有需要再加，过高的权重会出鬼图。 正面提示词：score_9,score_8_up,score_7_up,（必须加上）负面提示词：score_6,score_5,score_4, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry 免责声明：您基于本模型生成的文本内容由您自行维护并对其独立判断后使用，您需在遵守适用法律法规条件下使用，基于本模型生成内容产生的任何知识产权问题由您自行处理，您需对生成内容负责，本人对由此造成的任何损失不负责任。

Files with SHA256: 32709ca87b9ad34cf421cf523132719a5c453084eb81996af574233214f30ca8