<h1 id="taihou-character-lora">Taihou — Character LoRA</h1><p>So, this is my <strong>Taihou LoRA</strong>, trained with the goal of accurately reproducing <strong>Taihou from Azur Lane</strong> while maintaining her recognizable character identity across different outfits, poses, compositions and environments.</p><p>The main focus of this LoRA was <strong>character likeness</strong>, especially the combination of her distinctive facial features, long black hair, red eyes, hairstyle, body proportions and overall visual identity.</p><p>The idea was basically:</p><p><strong>Taihou → recognizable character → flexible character interpretation</strong></p><p>while still preserving her characteristic appearance even when changing clothing, poses, lighting and environments.</p><h3 id="trigger-word">Trigger Word</h3><p><code>Taihou</code></p><h3 id="recommended-weight">Recommended Weight</h3><p>Around <strong>0.7–0.9</strong> is a good starting point.</p><p>Lower weights can give you more flexibility with clothing, poses and environments, while higher weights will push Taihou's characteristic features and overall likeness more strongly.</p><h3 id="base-model">Base Model</h3><p><strong>HassakuXLIllustrious</strong></p><h3 id="recommended-quality-prompts">Recommended Quality Prompts</h3><pre><code>masterpiece, 8k, highly detailed face, highly detailed eyes,
individual hair strands, realistic fabric details,
detailed clothing, detailed lighting</code></pre><h3 id="a-few-things-about-the-lora">A few things about the LoRA</h3><p>This LoRA was trained primarily for <strong>Taihou's character likeness</strong>, rather than locking her to a particular outfit. It is intended to work across different clothing styles, poses, environments and compositions while keeping her recognizable features intact.</p><p>It works particularly well when combined with prompts describing the <strong>clothing, materials, lighting, environment and pose</strong>, allowing you to create different interpretations of Taihou without losing her identity.</p><p>The character's distinctive <strong>long black hair, red eyes, facial features and characteristic proportions</strong> are important parts of the intended result.</p><p>I recommend experimenting with the LoRA weight depending on the checkpoint and prompt you're using. <strong>0.7–0.9</strong> is a good general range, but some setups may produce better results with slightly lower or higher weights.</p><p>Have fun with it.</p>

Files with SHA256: d83d296981e9b9731eb52beab02687a9f8e75c5f1023b14d9990ec77f5fa7e80