LoRA for indie VTuber <a target="_blank" rel="ugc" href="https://x.com/dolliehitsujida">Dollie Hitsujida</a>, with tags for 3 outfits.The model should be way less overcooked, allowing for more versatility with outfits and hairstyles.Tags:Base:<pre><code>DollieHitsujida, white hair, twintails, ahoge, horns, sheep ears, blue eyes, heart shaped pupils, hair ribbons, x hair ornament</code></pre>Normal<pre><code>normaloutfit, pink Cardigan, plaid skirt, pantyhose, loafers, neck ribbon</code></pre>Pajamas<pre><code>PJoutfit, pink cardigan, blue pajamas, sleeveless, shorts, odd socks</code></pre>Wolf Costume<pre><code>WolfOutfit, animal hoodie, hood up, light gray hoodie, paw prints, bare legs, white knee socks</code></pre>This Lora being trained on a dataset consisting mostly of 3D model screenshots may cause some issues, however I haven't found any thus far.

Files with SHA256: a4d71670e7d7140cce331c154181538b01396b5c5b7ce8388c04254ceb63cc3b