This is trained using images from Twitter, I find them better looking. If you zoom in on the artist's images, you'll notice the lines aren't clean; they look like they were drawn with a pen or marker. Well, I've tried many things, but I haven't been able to replicate that, and judging by the other loras here, no one else seems to have been able to. If anyone has any ideas on how it could be done, I'd be grateful.<pre><code></code></pre>

ratatatat74 style

Better dataset and no trigger word

Files with SHA256: 06d2bfd10ce7d1dfcc9cbf69e9dc092b67d931b4435b405d28c82a20b6a67a9b