<a target="_blank" rel="ugc" href="https://www.patreon.com/GtoneroProject">Donate Coffee for Gtonero</a>Not recommended for realistic modelsMain tag : Milking machine, Secondary tag : nipples, lactation, Pose tag : leaning forward,norm retention: 98.14% | std: 0.010 Training Powershell info :<ul><li>Image Train : 160</li><li>DIM/Alpha : 12/6</li><li>Train batch size : 2</li><li>resolution : 768</li><li>mix/save_precision = BF16</li><li>optimizer_type = AdamW</li><li>GPU : RTX 4060 16GB 160W [Training Load 110W]</li><li>Torch 2.5 + cu12.1 + Xformers + gradient_checkpointing</li><li>Vram used : 10.1GB</li><li>Speed : 1.58s/it</li></ul><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/6e762625-dd20-46d5-b95d-7cd23649a196/width=525/6e762625-dd20-46d5-b95d-7cd23649a196.jpeg" />

Milking machine

LoRA is a beta with reduced DIM. Increased the number of images trained, but the steps during training remain the sameUpdate in Beta<ul><li>No need to tag leaning forward and upper body</li><li>Sometimes use the upper body if the resulting image is distorted.</li><li>I'll probably add a step in v.1.2 and reduce the LR later</li></ul>

Files with SHA256: 28e4cb51e72d72938dbf175cb377dce34c1dc1af24b1afe3d0caefecf585257e