0xSoul 发表于 2023-6-7 22:09:16

[SD 2.1 768][TextualInversion]card-style a 2.1 embedding for the modern digital fantasy style. cardstyleV3

<p><strong>card-style</strong> is the token for versions 1 and 2. The new version 3 was trained from the ground up using the same dataset and has a new token, <strong>cardstyleV3</strong>.</p><p>All of the v2 and v3 images are using the v2-1_768-nonema-pruned.ckpt checkpoint.</p><p>Trained on 140+ images (<em>260+ for v2</em>) from a variety of modern digital artists for card and videos games such as Gwent and Magic the Gathering. All of the artists have a similar style which has let me get something from the 2.1 checkpoint I was previously unable to achieve. I also recommend using the 2.1 trained checkpoint <a target="_blank" rel="ugc" href="https://civitai.com/models/5723/ultra-cmodel">Ultra Cmodel</a>. I suspect it used similar training images or it's just this modern style which everyone seems to be using. I also tested it with the <a target="_blank" rel="ugc" href="https://civitai.com/models/6465/prmj">PRMJ checkpoint</a>, the final image sample was made using it.</p><p></p><p>Update: A 1.5 SD version was requested. You can find it here: <a target="_blank" rel="ugc" href="https://civitai.com/models/9181/cardstyle15-card-style-for-the-15-checkpoint">https://civitai.com/models/9181/cardstyle15-card-style-for-the-15-checkpoint</a></p>
页: [1]
查看完整版本: [SD 2.1 768][TextualInversion]card-style a 2.1 embedding for the modern digital fantasy style. cardstyleV3