Please consider Buying Me A Coffee if you like my contributions to the community: <a target="_blank" rel="ugc" href="https://www.buymeacoffee.com/Rjox">https://www.buymeacoffee.com/Rjox</a><h3>TRIGGER WORD</h3>mrrpss<h3>TRAINING INFO</h3>Trained on 71 images 1492 steps 1 epoch<h3>RECOMMENDATIONS</h3>It works really well for txt2img pussy closeups, or inpainting. Wider/full-body shots are a mess.txt2img: Weight of 1.0. (When using a lower weight, it seems to mix the LoRA model with the checkpoint and you get weird stuff like 3 pussy holes. Not that you can ever have too many pussy holes.)Inpainting: Weight of 1.0 and a low denoising strength (0.2-0.4 or so). Your mask should have a decent mask blur amount, but that number depends on the image resolution. Just for example, for a 512x768 image, maybe 12+ mask blur. For 1024x1536, a mask blur of maybe 20+.<h3>ADDITIONAL RECOMMENDATIONS</h3>Other words/phrases that will help guide it:<ul><li>from behind (like ass-to-camera, doggy style)</li><li>front (facing the camera but legs not completely together, labia slightly spread)</li><li>spread (when legs are spread, regardless of angle/position)</li><li>closed (legs completely together from the front, labia closed)</li><li>naked woman (...or something like that. This isn't trained into the LoRA, but having more prompts like this will help it to know what it's looking at and where to place things)</li></ul>

Good Pussy

v1 was trained at 512x. upcoming v2 will be 768x

Files with SHA256: 49f4215a42816ea5f79a23b265170be9909fb947a331c3d489ce9c0686d2c927