Viggle AI launches JST-2, a physics-native 3D world model that builds scenes in seconds
Roger_M_Taylor · x · 2026-10-21
Viggle AI released JST-2, which it calls the world's first full-stack physics-native world foundation model.
- Unlike world models that learn in pixel/video space, JST-2 builds the entire stack in 3D from first principles: a text prompt or a photo yields a 3D scene in seconds, expandable in real time.
- Geometry stays consistent from every angle, and characters come with built-in motion, animatable from text or any video — all in one general architecture.
- For robotics, it means cheap on-demand 3D training worlds; the repost claims under a cent per scene.
Related event: Viggle Launches JST-2, a Physics-Native 3D World Foundation Model(13 posts)→
More from Embodied
- RoboARC shows robots get no free retries—unlike coding agents that can just rerun tests — ManlingLi_ · 2026-10-21
- All 20 robot arms failed in overcast weather at demo day, sparking low-level control debate — chris_j_paxton · 2026-10-21
- 'They still fail the burrito test': the case for putting AI models in bodies — generativist · 2026-10-21
- Three JEPA variants, one question: how well must a machine predict to plan? — TheTuringPost · 2026-10-21
- NVIDIA DGX Spark Founders Edition lands, creator preps local AI series — LinusEkenstam · 2026-10-21
- Shenzhen robotics founder meets Scobleizer: hardware-software fusion as the US-China bridge — Scobleizer · 2026-10-21