This is a LoRA for creating WYVERNS. I'm still having a bit of a problem with the <code>long range</code> trigger word, but overall, still pretty solid. Getting some nice variation, although it was a pretty hefty set of training data, too. The checkpoints still want to give the beasties 4 legs, so keep the base LoRA model around 1.0. I still seem some 3rd or 4th legs popping in at times.Like the STONE GIANT, I did use some incredibly basic prompts in the images just to make sure they were producing what was envisioned. I mean, <code>&lt;lora:RPGWyvern:1&gt; wyvern:1.2, mountains, realistic:1.3, 4k, best quality,</code> is pretty basic, right? And the <code>sitting</code> one, too is pretty basic.The images were created with hires.fx x2, denoising 0.23, 4x_foolhardy_Remacri upscalerThe trigger words for this model are the following:<pre><code>blurry, claws, fangs, flying, green body, green wings, horns, long range, open mouth, orange eyes, red eyes, red wings, scales, sharp teeth, sitting, solo, tail, teeth, wings, wyvern, yellow eyes, yellow wings</code></pre>

Files with SHA256: 70217ab905b218aee09a70a75c13e548d4b56f2beefbd5a11a88904b104e6343