Slap (and Self Slap)
LoRA for generating slaps
1 min read
A new LoRA (a small add-on to a base model) for the MiniMax H3 generator that introduces slapping and self-slapping concepts into video. The model is trained to understand natural language via trigger words, letting you generate shots where a character slaps themselves in the face or gets struck by an offscreen hand. Starting with a weight of 0.8 is recommended, lowering it to 0.5–0.6 for high-motion scenes. It is well suited for content creators experimenting with unusual physics and actions in video generation.
Key features
- Adds slapping and hitting concepts to video
- Built specifically for the MiniMax H3 base model
- Supports control via natural language text prompts
- Flexible weight tuning to control action intensity
What you can do
- Generate a video where a character suddenly slaps themselves in the face
- Create a scene with a slap coming from an unseen hand offscreen
- Add dynamic and emotional interaction elements to your project
Example
Use trigger words in your text prompt: She slaps herself in the face. or She is slapped in the face by a hand from offscreen.
Read more
DetailsBase model: MiniMax H3
Adapter type: LoRA
Recommended weight: 0.8 (or 0.5–0.6 for complex scenes)














Comments