✦
Summary
Viggle JST-2 is a physics-native world foundation model for generating scenes, characters, and their motion. It represents persistent matter as 3D atoms with explicit positions and learned attributes and dynamics, rather than representing the world only as independent video frames.
Key Features
Persistent 3D atom representation.
Joint space-and-time modeling.
Image-to-world encoding.
Prompt-conditioned world generation.
Viewpoint-conditioned output.
Motion and force conditioning.
3D Gaussian-splat outputs.
Developer API access.
About Viggle JST-2
It supports interactive content and world-building workflows that need coherent objects as viewpoints and motion change. Core capabilities include Persistent 3D atom representation, Joint space-and-time modeling, Image-to-world encoding, and Prompt-conditioned world generation.
Use the provider's product or developer platform; account access and the available usage plan determine deployment options. Additional characteristics include Viewpoint-conditioned output, Motion and force conditioning, 3D Gaussian-splat outputs, and Developer API access.