<h2 id="heading-473">I would highly recommended using both of trigger words, u usually will get better results.</h2>Also, couple of recommendations:<ul><li>Use lawlassYiffymix20 for results as showed on the page</li><li>recommended clip skip 2</li><li>set around 0.9 scale (I was using exactly 0.9)</li><li>the model trained on 512x512 images</li></ul>If you dont get good results, write an feedback with your generations. I'll try to fix this.

Cream the rabbit (LoRA)

<ul><li>More epochs</li><li>More images ai to eat</li><li>release v2</li></ul>

Files with SHA256: faf8dff752e3a300a914c6f37d1f5f7bb4ce438dde833a573565d9629451346b