### 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](https://huggingface.co/Viggle/Viggle-Animate) ·
[Demo](https://huggingface.co/spaces/Viggle/viggle-animate) ·
[Write-up](https://viggle.ai/research/viggle-animate-character-replacement-from-a-repainted-frame?utm_source=huggingface&utm_medium=social&utm_campaign=org-card)
|
### 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](https://huggingface.co/Viggle/Meridian) ·
[Demo](https://huggingface.co/spaces/Viggle/Meridian) ·
[Write-up](https://viggle.ai/research/meridian-a-new-perspective-on-space-and-time?utm_source=huggingface&utm_medium=social&utm_campaign=org-card)
|