Odyssey-3 world model claims unified control of robots, drones and cars with minimal task data
SucceededMind · x · 2026-09-16
Odyssey unveiled Odyssey-3, a foundation world model meant to replace per-machine control systems. Trained on visual observation, it reportedly grasps gravity and momentum from broad pretraining and maps that knowledge to new hardware with minimal task data — covering driving (on real roads in India), drone piloting, robot arms, synthetic agent training, and even video games. The team calls it a shift from isolated control systems to unified physics engines, with public rollout weeks away.
Related event: Odyssey-3 World Model Controls Robots, Cars and Drones(9 posts)→
More from Embodied
- 7 hours of human data beats 17 hours of public datasets at 85% robot task success — YuXiang_IRVL · 2026-09-16
- One prompt to a real object: AI agent models in Blender and 3D-prints the result — anthara_ai · 2026-09-16
- NVIDIA releases FoundationPose on Hugging Face: unified 6-DoF pose model, no fine-tuning needed — _akhaliq · 2026-09-16
- Chinese Robots Now Building Robots: New Factory Produces a Humanoid Every 10 Minutes — Polymarket · 2026-09-16
- Ex-Meta CTO invests in Planted, robots that make solar installs 2x denser — sudoraohacker · 2026-09-16
- Apptronik and Agility on robot safety: the path home runs through industry — KyleMorgenstein · 2026-09-16