A LoRA focused on bowlegged pose for Anima, trained on Anima Base v1.0 with Citron Legacy.Trained with 474 images, 10 epochs, and 15400 total training steps.Trigger word: <code>bowlegged pose</code>Tips:<ul><li><code>legs apart</code> gives the regular bowlegged stance.</li><li><code>wide spread legs</code> pushes the legs much farther apart.</li><li><code>low stance</code> lowers the center of gravity without turning it into a full squat.</li><li><code>feet turned outward</code> and <code>on tiptoes</code> work great depending on the pose.</li><li>Works nicely with different camera angles, clothing, and art styles.</li><li>Also supports sitting, leaning, back views, prone, supine, and even some interaction poses.</li><li> I intentionally included a lot of pose variations instead of only straight standing references, so it should stay pretty flexible instead of locking itself into one exact composition.</li></ul>Hope you enjoy it!Keep gooning y'all~Training Data ↓Base Model: Anima Base v1.0 Network Dim: 20 Network Alpha: 20 Learning Rate: 1e-4 Optimizer: AdamW8bit Resolution: 768 Batch Size: 1 Epochs: 10 Steps: 15400 Caption Dropout: 0.1 Mixed Precision: bf16Dataset:<ul><li>474 manually curated images</li><li>1540 effective images per epoch</li><li>15400 total training steps</li><li>Multiple pose variations (standing, low stance, sitting, leaning, back view, prone, supine, interactions and 3D references)</li></ul>

Files with SHA256: e19cfd2b58c9d158670b28ce698aacbd6136b773b58d8920b5727fe4c4910bcb