A Lora designed to generate highly detailed hand-over-mouth scenes.Illustrious v1, Pony v1, Pony v2 are deprecated!<h2 id="prompting-(illustrious-v3):">Prompting (Illustrious V3):</h2><ul><li>The main trigger is "handgag" or "handsmother" (the hand will likly also cover the nose)</li><li>Dynamics: Supports all four character pairings: "female on female", "male on female", "female on male", and "male on male". Can generalize well if you use "anthro on female", or "monster on male", etc.</li><li>View Angles: Works across "front view", "3/4 view", and "side view".</li><li>Hand Detail: Easily customize the hand with descriptors like "small hand" "big hand", or specific nail colors/lengths ("red nails", "long nails" etc.).</li><li>Gloves: Use the trigger "gloved" for a wide variety of gloves, including specific terms like "leather glove," "black glove," or "fingerless glove"</li></ul><h2 id="tips:">Tips:</h2><ul><li>Follow the checkpoint's recommendations for steps, sampler, etc.</li><li>Use ADetailer to improve the details of the hands, nails, or jewelry.</li></ul><h2 id="prompting-(reverse-grip):">Prompting (Reverse grip):</h2><ul><li>The main trigger is "handgag", and the two persons in the scene: "female to female", "male to female", "female to male", "male to male".</li><li>To describe the person covering the mouth, add "dom" after the tag. Examples: "ginger handgagger, blue eyes handgagger, red jacket dom" etc.</li><li>The same goes for the victim, such as "white shirt sub, brown hair sub, surprised sub" etc.</li><li>You can also describe the hand with "small hand, big hand, strong hand, red nails, blue nails, long nails" etc.</li></ul>The model also supports gloves; the main trigger is "gloved". But you can also describe the glove like "leather glove, black glove, fingerless glove", etc.<h2 id="prompting-(illustrious-v2):">Prompting (Illustrious V2):</h2>Outdated!<ul><li>The main trigger is "handgag", and the two persons in the scene: "female on female", "male on female", "female on male", "male on male".</li><li>To describe the person covering the mouth add "dom" after the tag. Examples: "ginger dom, blue eyes dom, red jacket dom" etc.</li><li>The same goes for the victim, such as "white shirt sub, brown hair sub, surprised sub" etc.</li><li>You can also describe the hand with "small hand, big hand, strong hand, red nails, blue nails, long nails" etc.</li><li>The model also supports gloves, the main trigger is "gloved". But you can also describe the glove like "leather glove, black glove, fingerless glove", etc.</li><li>The dataset includes various poses, view angles, environments, and styles, so you will not struggle to create what you want.</li></ul><h2 id="limitations-and-biases">Limitations and biases</h2>The model is not perfect:<ul><li>(Reverse grip only) : Small dataset and unknown pose from all models, expect a lower success rate (~70%).</li><li>Struggles with realistic images. I do not recommend using this model on realistic checkpoints.</li><li>The person covering the mouth will often be taller than the victim.</li><li>The risk of an extra finger or a missing finger is about 20% with Pony and 10% with Illustrious.</li></ul>Please use this LoRA responsibly, following ethical and moral values. I take no responsibility for its misuse.

Handgag (Anima/Illustrious)

Just 100x better than the first version.

Pony_handgag_V2.safetensors

Pony v2

Anima (Prev 3)

Illustrious v3

Illustrious v2

Illustrious (Reverse) v1

Illustrious v1

Pony v1

This LoRA reproduces decent hand over mouth scenes.The model currently has some limitations:<ul><li>There is only one pose (from behind).</li><li>For NSFW content, it needs more tags or slightly increased NSFW tag weights.</li></ul>But now the good things:<ul><li>V2 has a success rate of approx. 80% (tested with AutismMix confetti)</li><li>No overfit, any creature can cover the mouth of any creature.</li><li>Supported descriptions for the person covering the mouth and the victim.</li></ul>Prompting tips:<ul><li>The&nbsp;main trigger&nbsp;is "handgag", and the two persons in the scene: "female to female", "male to female", "female to male", "male to male".</li><li>To&nbsp;describe the person&nbsp;who is covering the mouth add "handgagger" after the tag. Examples: "ginger handgagger, blue eyes handgagger, red jacket handgagger" etc.</li><li>The same goes for the victim, such as "white shirt victim, brown hair victim, surprised victim" etc.</li><li>You can also&nbsp;describe the hand&nbsp;with "small hand, big hand, strong hand, red nails, blue nails" etc.</li><li>The model also supports&nbsp;gloves, the main trigger is "gloved". But you can also describe the glove like "leather glove, black glove, fingerless glove", etc.</li><li>The dataset includes various environments and humanoids, so it won't struggle to create robots, furry, aliens, ...</li></ul>Use tips:<ul><li>20 to&nbsp;30 steps max. Too many steps will result in increased risks of extra fingers.</li><li>FaceDetailer&nbsp;can improve the thin details of the hand like nails, rings, etc.</li><li>Using a&nbsp;tiled upscaler&nbsp;will produce better results.</li></ul>Please use this LoRA responsibly following ethical and moral values. I take no responsibility for its misuse.

Files with SHA256: e91ee161cc6194c7762e938a466526f86654e16ead26b06d7dc271d69e4f834a