Are you tired of having an overfit model for every type of mummification? I've got your back.This all-in-one model is trained for mummification or cocooning with almost any material, including web, bandages, vines, slime, latex, tape, and straitjackets. It can also generalize well to other materials.<h2 id="prompting:-e0kf9ltz6">Prompting:</h2><ul><li>Tags for selecting the mummification material: tape bondage, web bondage, vines bondage, slime bondage, latex bondage, bandage bondage, ice bondage, straitjacket bondage, etc.</li><li>Tags for selecting the material color: tape + color or web + color, etc. (Examples: red tape or pink web, etc.)</li><li>Tags for exposing a part of the body: exposed breasts, exposed hips, exposed ass, exposed feet, exposed head, exposed eyes, exposed hair, etc.</li><li>Tags for gags: tape gag, silk gag, vines gag, bandage gag, etc.</li><li>Choosing the thickness of the mummification: 'thin cocoon' will leave clear shapes on the body, 'thick cocoon' will make a large cocoon around the body. And 'translucent cocoon' for a... translucent cocoon.</li><li>Adding a dom: You can add a not-mummified character by using the tag 'activate_dom'. Describe the dom by adding 'dom' after all relevant tags. Example: male dom, black hair dom, white shirt dom, etc.</li><li>Recommended pos: 'detailed background'; Since many images in the dataset have simple backgrounds, I added regularization images tagged to improve the background if needed.Recommended neg: 'low quality'; some pics in the dataset are slightly below native resolutions.</li></ul><h2 id="limitations-and-biases:-z572xfn07">Limitations and biases:</h2><ul><li>The model will tend to generate female characters. But if you use different tags related to males or increased weights for male tags, it will be fine.</li><li>The success rate of a good pic when using 'activate_dom' is lower but far better than in v1.0</li></ul>Please use this LoRA responsibly, following ethical and moral values. I take no responsibility for its misuse.

Files with SHA256: a1e0ba2490405298b8250458312f0a2bea478edbf31c86c052a6ddd46a14bbba