This is more of a character pack than a style, but I found better results using the style tag rather than the character tag during LoRA training. You may notice some issues with backgrounds or occasional deformations. If any generation issues are consistent, please let me know so I can address them in the next version. <span style="background-color:rgb(26, 27, 30);font-family:-apple-system, &quot;system-ui&quot;, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">The characters are listed in order based on how much training data they were given. For example, Yasmin, near the bottom, probably only had about 3 photos, whereas Brittany probably had 80, so keep that in mind. Enjoy!IMPORTANT NOTE: All of my models were trained and tagged using Danbooru-style tagging conventions (e.g., <code>courtney_(fellatrix)</code>, <code>bob_cut</code>, <code>medium_hair</code>, etc.)

Fellatrix XL

Same notes as the pony model. This is a test model, so I’ll be updating it later. I’d appreciate any feedback on visual glitches or persistent generation issues.

Files with SHA256: daea3d4f6e43debc27d8e1773856f15a1fd67f60e2c0774a7bf422b428223bd8