<p><strong>Preview images do not use other LoRAs.</strong></p><h1 id="commercial-use-of-this-model-is-not-allowed."><span style="color:rgb(250, 82, 82)">❌💲Commercial use of this model is <strong>not</strong> allowed.</span></h1><h3 id="i-trained-this-lora-on-my-own-art-only-for-this-reason-it-has-stricter-permissions-than-my-other-loras."><span style="color:rgb(250, 82, 82)">I trained this LoRA on my own art only, for this reason it has stricter permissions than my other LoRAs.</span></h3><h3 id="this-model's-dataset-contains-nsfw-materials."><span style="color:rgb(250, 176, 5)">🔞This model's dataset contains NSFW materials.</span></h3><p></p><p>The PonyXL V0.1 version was trained on only 28 images with rushed tagging. The PonyXL V1 version is the first one I consider "proper".</p><p>None of the pictures in the datasets had watermarks/signatures/artist names (although a few have text), if you get them - they're NOT from this LoRA's dataset. There may be bias towards certain elements.</p><p>It may come off as ironic, but I'm asexual. This is why I don't draw spread vulvas correctly (I tried using a SCHEMATIC as a reference, and I still felt uncomfortable), and I may mess up penises too. Whatever this model spits out is probably better than what I can do myself.</p><p></p><h3 id="issues">⚠ISSUES</h3><p>There seems to be a risk of multiple/conjoined worms, I suggest adding "multiple worms" into negatives and strengthening "solo" (assuming you prompt for solo). No promises.</p><p>A single picture of a worm with a more realistic body seems to pollute the results, especially anthrofied ones. I'll remove it if I update this.</p>

Files with SHA256: c47678ba4b973a97e857225d57bd2b343915a9f173884b89df6493a53e1a6930