<s>Trained using only 7 similar images, so it's somewhat rigid with lighting.</s> (Fixed in V2)<h1 id="tags:-n1hanue2h">Tags:</h1><h3 id="appearance:-e8nc5wzbp">Appearance:</h3><pre><code>2dmita, 1girl, solo, blue eyes, blue hair, short twintails, medium hair, red bow, hair bow, hairclip,</code></pre><h3 id="outfit:-p02ysyyib">Outfit:</h3><pre><code>red choker, ribbon choker, red shirt, short over long sleeves, striped sleeves, pleated skirt, shirt tucked in, miniskirt, blue skirt, red socks, kneehighs, </code></pre><h3 id="add-to-negative:-3zfhwn03b">Add to negative:</h3><pre><code>bloom, glowing, </code></pre>

2D Mita - MiSide

Trained using Regularization (using same dataset) and it has considerably improved flexibility of LORA, backgrounds and lighting far better, but somewhat less style accurate compared to V1.

2DMita_IL_V2.safetensors

V2 (Illustrious-XL)

V1 (Illustrious-XL)

Trigger Words:2dmita, 1girl, solo, blue eyes, blue hair, short twintails, medium hair, red bow, hair bow, hairclipOutfit:red choker, ribbon choker, red shirt, short over long sleeves, striped sleeves, pleated skirt, shirt tucked in, miniskirt, blue skirt, red socks, kneehighsAdd to negative:bloom, glowing

2D Mita - MiSide.safetensors

v1.0

Creator: <a target="_blank" rel="ugc noopener noreferrer nofollow" href="https://civitai.com/user/MV_QT">MV_QT</a>You can support the creator here: <a target="_blank" rel="noopener noreferrer nofollow" href="http://boosty.to/mv_qt/donate">boosty.to/mv_qt/donate</a>Trigger Words:2dmita, 1girl, solo, blue eyes, blue hair, short twintails, medium hair, red bow, hair bow, hairclipOutfit:red choker, ribbon choker, red shirt, short over long sleeves, striped sleeves, pleated skirt, shirt tucked in, miniskirt, blue skirt, red socks, kneehighsAdd to negative:bloom, glowing

Files with SHA256: 24441400bd4f0fc8376d700f92813666f1a0b3677b0a84486548c64fa230f51a