This model was trained using images from nonco for NoobAI-XL (0.5 epsilon). Trigger word <code>nonco</code> is optional, but it does make the style more pronounced.I really like nonco's style and anatomy. I was going to release this for pony initially, but then <s>illustrious</s> noobai came out. This is the same dataset as the <a target="_blank" rel="ugc" href="https://civarchive.com/models/860133/nonco-style-dora-illustrious-xl">Illustrious-XL 0.1 DoRA</a>.Note: Gens on civit look borked, you might have to reduce strength. I use forge, and it doesn't have issues with the strength.Changelog:<ul><li>v9 uses 57 color images (manga &amp; booru) and 26 black&amp;white images (manga); labelled with wd1.4 and manually cleaned</li></ul>Strength plot (<a target="_blank" rel="ugc" href="https://files.catbox.moe/vfhvh4.png">1</a>, <a target="_blank" rel="ugc" href="https://files.catbox.moe/x7hz5w.png">2</a>)<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/77c0e4ce-0882-425d-ad12-13cc96ec8c32/width=525/77c0e4ce-0882-425d-ad12-13cc96ec8c32.jpeg" /><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/3647e908-6449-4eb5-894c-a40289dde632/width=525/3647e908-6449-4eb5-894c-a40289dde632.jpeg" />

nonco style DoRA (NoobAI-XL)

DoRA trained on 83 images with kohya_ss (NoobAI-XL 0.5 epsilon)<ul><li>optimizer - AdamW</li><li>lr - 2e-4</li><li>unet lr - 1.5e-4</li><li>text lr - 4-e4</li><li>lr scheduler - cosine</li><li>gradient accumulation - 4</li></ul>

Files with SHA256: 93c32fb08cd19808136ddb7e74290dce860ba17c2c330edf40bccdd308976aa5