README.md

metadata

title: Viggle AI
emoji: 🕺
colorFrom: indigo
colorTo: purple
sdk: static
pinned: false

Viggle AI builds Spatial GenAI. Our models generate character, motion and scene together in one shared 4D space — the foundation under Meme Maker, PINOC, Viggle Games and the Viggle API, used by 50M+ people.

This organization is where we open-release models. Weights, inference code and a runnable demo Space for each one.

Open releases

Viggle-Animate Character replacement in video. Repaint one frame of a clip in any image editor; the model propagates that edit across the shot with motion, camera and timing untouched. No pose skeleton, no segmentation mask, no text prompt. A 33 B finetune of MiniMax-H3, DMD-distilled to four sampling steps — 124 frames in 26 s on one B200. Weights · Demo · Write-up

Meridian A new perspective on space and time. Revisit footage you already have from a camera path you author: orbit the action, hold one moment still, or follow a retimed sequence — and from a single photograph too. Geometry comes from the video itself, so you can preview the framing before you generate. Takes of 3–10 s at 24 fps. Weights · Demo · Write-up

Elsewhere

viggle.ai · Research · API docs · API portal · Discord · X · YouTube