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>You can run WAI-ANI-NSFW-PONYXL and use its API on SinkIn:<a target="_blank" rel="ugc" href="https://sinkin.ai/m/2zgReX9">https://sinkin.ai/m/2zgReX9</a><hr /><h1 id="v14-released.-0dd7br8zl">V14 released.</h1>Increased composition diversity and improved model accuracy now allow images to be generated with fewer steps.<h2 id="recommended-setting-kqfs7j4n7">Recommended Setting</h2><h3 id="steps:20-30-a2841j4w3">Steps:20-30</h3><h3 id="cfg-scale:-5-7-hge3gwra1">CFG scale: 5-7</h3><h3 id="sampler:-euler-adpm++-2m-karras-7gc2n7sm9">Sampler: Euler a/DPM++ 2M Karras</h3><h3 id="adetailer-face_yolov8ns.pt-use-can-fix-eyes-9wc2h6jcw">ADetailer face_yolov8n/s.pt use can fix eyes</h3><h3 id="for-the-example-images-qds2h65da">For the example images</h3>I used 1024x1360,generated directly without AD or hires fix.Positive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>worst quality,bad quality,jpeg artifacts, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark,</code></pre><hr /><h1 id="v13-released.-poprjihzr">V13 released.</h1><ul><li>Increased body stability and accuracy.</li><li>overall balance adjustment.</li></ul><h2 id="recommended-setting-u0qh3zebz">Recommended Setting</h2><h3 id="steps:-30-wv1nst4jf">Steps: 30</h3><h3 id="cfg-scale:-7-ait3luw3w">CFG scale: 7</h3><h3 id="sampler:-euler-adpm++-2m-karras-o6o23533p">Sampler: Euler a/DPM++ 2M Karras</h3><h3 id="adetailer-face_yolov8ns.pt-use-can-fix-eyes-g62djd3n6">ADetailer face_yolov8n/s.pt use can fix eyes</h3><h3 id="for-the-example-images-ai92tf9rz">For the example images</h3>I used 1024x1360,generated directly without AD or hires fix.Positive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>worst quality,bad quality,jpeg artifacts, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark,</code></pre><hr /><h1 id="v12-released.-rf2308ahd">V12 released.</h1><ul><li>Style adjustment, background adjustment, and increased stability.</li><li>Added more compositions.</li></ul><h2 id="recommended-setting-vor7cury3">Recommended Setting</h2><h3 id="steps:-30-forgzmpw8">Steps: 30</h3><h3 id="cfg-scale:-7-lvvc2hvsl">CFG scale: 7</h3><h3 id="sampler:-euler-adpm++-2m-karras-82o2n084x">Sampler: Euler a/DPM++ 2M Karras</h3><h3 id="adetailer-face_yolov8ns.pt-use-can-fix-eyes-ptr9gwc6h">ADetailer face_yolov8n/s.pt use can fix eyes</h3>Positive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>score_6, score_5, score_4, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark,</code></pre><hr /><h1 id="v11-dmmzd6ilc">v11</h1><h3 id="-r28kaeyl4"></h3><h3 id="weight-adjustment-balancing-the-art-style-mdijozia6">Weight adjustment, balancing the art style</h3>All example images(Except the cover) are generated at 1024x1360, without using AD and hires fix.<h2 id="recommended-setting-0yvhg3h0r">Recommended Setting</h2>Steps: 30CFG scale: 7Sampler: Euler a/DPM++ 2M KarrasADetailer face_yolov8n/s.pt use can fix eyesPositive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>worst quality,bad quality,jpeg artifacts, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark, </code></pre><hr /><h1 id="v10-2bajuxpz1">V10</h1><ul><li>Better Background</li><li>Clothes details up</li><li>Better mature female</li></ul><h2 id="recommended-setting-m1vaq9cng">Recommended Setting</h2>Steps: 30CFG scale: 7Sampler: Euler a/DPM++ 2M KarrasADetailer face_yolov8n/s.pt use can fix eyesPositive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>worst quality,bad quality,jpeg artifacts, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark, </code></pre>++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<h2 id="v9-hyper12step-0k6r0qsl6">V9 Hyper12step</h2><h3 id="please-refer-to-the-instructions-on-the-right-side-of-v9-hyper12step-for-usage.-9qp2vzfbk">Please refer to the instructions on the right side of V9 Hyper12step for usage.</h3><h1 id="v9-released.-2m4qfkrww">V9 released.</h1><h3 id="v8-greaterv9-vo47a7xib">v8-&gt;v9</h3><ul><li>adjusted the facial data to make it more versatile, allowing for easier generation of faces across various ages.</li><li>added more NSFW materials.</li><li>attempted to reduce the generation of some unnecessary details.</li></ul><h2 id="-5iwvq1ujq"></h2><h2 id="recommended-setting-vp04o1mlv">Recommended Setting</h2>Steps: 30CFG scale: 7Sampler: Euler a/DPM++ 2M KarrasADetailer face_yolov8n/s.pt use can fix eyesPositive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>worst quality,bad quality,jpeg artifacts, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark, </code></pre>__________________________________________________________________________________________________<h3 id="hyper12step-released.-5lioxxzfd">hyper12step released.</h3>about sizeVarious combinations from 768x768 to 1360x1360 can produce good images.Recommended SettingSteps: 12CFG scale: 3.5Sampler: Euler aNot very necessary to use ADetailer correctionPositive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>score_6, score_5, score_4, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark,</code></pre><h3 id="ps.the-example-images-use-a-size-of-896x1192.-no-ad-no-hires.-fix-soqzlyiuy">PS.The example images use a size of 896x1192. no AD no Hires. fix</h3>__________________________________________________________________________________________________<h3 id="version-8-released.-cdrr4xcso">Version 8 released.</h3>v7-&gt;v8<ul><li>Added some training related to camera angles and backgrounds.</li><li>Added more NSFW materials for additional training.</li><li>Added some training for special effects.</li><li>Optimization of overall composition logic.</li><li>Fixed some minor issues.</li><li>better eyes and hands.</li></ul><h3 id="recommended-setting-h3xiwnufc">Recommended Setting</h3><h1 id="same-as-v7-9x5tr5gm8">Same as v7</h1>__________________________________________________________________________________________________<h3 id="version-7-9yvvvzl0o">Version 7</h3>v6-&gt;v7<ul><li>Fixed the issue where images appeared entirely covered with snowflakes.</li><li>Revised the composition logic to be more suitable for general scenarios.</li><li>Added new assets for backgrounds, mature women, and clothing.</li><li>Ensured consistency with the previous model and further fixed the issue of blurry eyes in medium and long shots.</li><li>Fixed some minor issues.</li></ul><h1 id="v7-recommended-setting-x2ap9wrt8">V7 Recommended Setting</h1><h3 id="steps:-30-k55h036h4">Steps: 30</h3><h3 id="cfg-scale:-7-1aw715qhk">CFG scale: 7</h3><h3 id="sampler:-euler-adpm++-2m-karras-xt48l5suu">Sampler: Euler a/DPM++ 2M Karras</h3><h3 id="adetailer-face_yolov8ns.pt-use-can-fix-eyes-4ej2ph4rr">ADetailer face_yolov8n/s.pt use can fix eyes</h3>Positive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,</code></pre>Negative Prompt<pre><code>score_6, score_5, score_4, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark,</code></pre>________________________________________________________________________________________________Regarding the 8step model, it is a lightweight model that can generate images in as few as 8 steps. However, the quality and text comprehension will be relatively lower. If speed is not a major concern for you, it is recommended to continue using version 6.0.<h3 id="8-step-setting-qw3s638uu">8 STEP Setting</h3>Steps: 8-12CFG scale: 3Sampler: Euler aADetailer face_yolov8n/s.pt use can fix eyesPositive Prompt and Negative Prompt same 6.0__________________________________________________________________________________________________<h3 id="6.0-t5dizlz1j">6.0</h3><h3 id="v5-greaterv6-change-point-pemeolnrq">v5-&gt;v6 Change point</h3><ul><li>Better Background</li><li>better eyes</li><li>fix vae</li></ul><h1 id="recommended-setting-yqn47j4s2">Recommended Setting</h1><h3 id="steps:-30-vj1c8s575">Steps: 30</h3><h3 id="cfg-scale:-7-7rxotcvfk">CFG scale: 7</h3><h3 id="sampler:-euler-adpm++-2m-karras(better-eyes)-pu6yrkipe">Sampler: Euler a/DPM++ 2M Karras(better eyes)</h3><h3 id="adetailer-face_yolov8ns.pt-use-can-fix-eyes-39l0e6jub">ADetailer face_yolov8n/s.pt use can fix eyes</h3>Positive Prompt<pre><code>score_9, score_8_up, score_7_up,source_anime,BREAK</code></pre>Negative Prompt<pre><code>score_6, score_5, score_4, source_cartoon, 
3d, (censor),monochrome,blurry, lowres,watermark,</code></pre><h3 id="you-can-view-the-supported-art-styles-from-https:civitai.comarticles5715-vbflweqoa">You can view the Supported art styles from <a target="_blank" rel="ugc" href="https://civarchive.com/articles/5715">https://civarchive.com/articles/5715</a></h3><h3 id="thank-you-deepdark_fantasy514-for-providing-the-test.-1wfpzltwt">Thank you, DeepDark_Fantasy514, for providing the test.</h3><h3 id="-y8h3kh9k3"></h3>v4-&gt;v5 Change point<ul><li>Better Background</li><li>Clothes details up</li><li>Add some mature female</li><li>Little better hands and eyes</li></ul>Recommended Setting<ul><li>Steps: 30</li><li>CFG scale: 7</li><li>Sampler: Euler a</li><li>ADetailer face_yolov8n/s.pt use can fix eyes</li></ul>Positive Prompt<pre><code>score_9, score_8_up, score_7_up,</code></pre>Negative Prompt (THX for snipsnapsnipsnap recommend )<ul><li>Base:score_6, score_5, score_4, source_cartoon</li><li>Optional: source_furry, source_pony can bleaches skin</li><li>else: 3d, (censor),monochrome,blurry, lowres,watermark,</li></ul>I often use it now[score_6, score_5, score_4, source_cartoon,3d, (censor),monochrome,blurry, lowres,watermark, ]

WAI-ANI-PONYXL

waiANIPONYXL_v8Hyper12step.safetensors

V8_hyper_12step

v14.0

v13.0

V9_hyper_12step

v9.0

v8.0

v7.0

8step

v6.0

v5.0

v4.0

v3.0

v2.0

v1.0

Files with SHA256: f897355eb6485a0e90fbf382f0331196e3127d964859ec05f6c16de4c05689ea