Odyssey-3 world model: driving policy trained on just 20h of simulation hits real Indian streets
rohanpaul_ai · x · 2026-09-16
Odyssey unveiled Odyssey-3, a foundation world model targeting robotics' biggest bottleneck: task-specific data. It pretrains a broad world model, then attaches an action decoder trained on only tens of hours of demonstrations. Reported data needs are unusually small: tens of hours for robot arms and humanoid teleoperation, and a driving policy trained on just 20 hours of simulation drove real Indian streets. It also powers drones, AI training, and video games. Numbers are early, but the thesis is that world-model pretraining can absorb what every downstream policy would otherwise relearn.
Related event: Odyssey unveils general-purpose world model Odyssey-3 in early preview(13 posts)→
More from Embodied
- ANYbotics begins customer deliveries of ANYmal X, first legged robot certified for explosive atmospheres — lukas_m_ziegler · 2026-11-02
- RAM prices tripled, so Tesla cut a third of memory from Optimus's brain — aakashgupta · 2026-11-02
- Too lazy on weekends? Watch this robot tidy up a desktop — flngr · 2026-11-02
- Wearing Whoop and Apple Watch on a single dual-slot band — tinyfool · 2026-11-02
- REA: async brain-cerebellum runtime for robots where policies never wait on the LLM — Pengfei Zhou · 2026-11-02
- Robot Combat Sports Coming to Japan with a Permanent Arena in Tokyo — cixliv · 2026-11-02