<a target="_blank" rel="ugc" href="https://ko-fi.com/gtoneroproject">Donate Coffee for Gtonero</a>Main tag : Suspension, Important tags <pre><code>arms behind back, bound, shibari, bound legs arms, </code></pre>Optional tag<pre><code>knee down, folded, legs behind head, spread legs, </code></pre>Training LoRA Powershell info :<ul><li>resolution = 768</li><li>Batch size = 2</li><li>network_dim = 32</li><li>network alpha = 32</li><li>mix/save_precision = FP16</li><li>optimizer_type = Radam</li><li>optimizer_args_other = NONE</li><li>GPU RTX 4060Ti 16GB</li><li>Torch 2.7+cu128 + Xformers + gradient_checkpointing</li></ul>

Bondage Suspension

It's still not perfect. may need to filter the image training model again.

Files with SHA256: 2da9f4de695b200d83c4ea261d0fd5f1eeed9e3e856a6f90d309e2f986a5fd65