Viggle unveils JST-2, an AI that builds physics-enabled 3D worlds in seconds
141_1337 · reddit · 2026-10-20
Viggle has announced JST-2, a generative model that, unlike typical AI video generators, builds environments directly in 3D:
- Generates full 3D worlds in seconds and expands scenes in real time
- Physics is built in, so environments behave physically rather than being pixel-level video
- Also creates animated characters
The demo video shows the model in action, marking a notable step from video generation toward interactive 3D scene generation.
Related event: Viggle Launches JST-2, a Physics-Native 3D World Foundation Model(7 posts)→
More from Multimodal
- Stunning black & gold AI portrait made with Recraft V4.1 flash — aziz4ai · 2026-10-20
- AI workflow turns ordinary photos into stunning hand-embroidery art — aziz4ai · 2026-10-20
- Anthropic engineer publishes the full prompts behind a dozen Opus 5.5 code-made films — EricBuess · 2026-10-20
- Live show visuals and DJ avatar fully generated in ComfyUI, synced to Ableton — PurzBeats · 2026-10-20
- Yeufonic: open-source frontend for YuE2/ComfyUI aimed at musicians, with local LoRA training — AcrobaticMaize2408 · 2026-10-20
- OmniXRef: a unified framework for multimodal multi-reference video editing — drsrinathsridha · 2026-10-20