<p>Lora trained in the DAGASI's art style.</p><p></p><h2 id="illustrious01-v2">Illustrious01 v2</h2><p>This has been my most used model by so far. So I decided remake it with a bigger dataset. The original one was really hard pruned for focus certain features. Now I increased the dataset around 20 times and I'm using a new training config. Is a bit experimental so I hope if you find any issue with it, please share it with me for fix it and improve next models.</p><p>Weight: .8 - 1</p><p>know issues:</p><ul><li><p>Due the increased furry dataset included in the new version, you can expect more issues with fingers number.</p></li></ul><h2 id="illustrious01-v1">Illustrious01 v1</h2><p>Weight: .8 - 1</p><h2 id="pony-v1">Pony v1</h2><p>Weight: .8 - 1</p>

DAGASI Style (Illustrious01 | Pony v6)

Files with SHA256: cb2698e58675fb15cc1f2ade95661434face256d054583b0fbc09a55bc91fb57