RewardAI's OM-1 Robot Foundation Model Zero-Shot Generalizes to Arms and Humanoids
zipengfu · x · 2026-09-15
RewardAI introduced OM-1, its first robot foundation model that zero-shot generalizes to any robot embodiment: table-top arms, industrial arms and humanoids.
- Learned directly from human manipulation data, with no teleop or robot data
- Close to human-level dexterity and efficiency
- Multi-robot collaboration supported
The reposter noted the demo video plays at 1x speed and looks remarkably smooth.
Related event: RewardAI Unveils OM-1, a Robot Foundation Model Trained on Human Data(45 posts)→
More from Embodied
- Developer Seeks Peers Working on Drones, Gaussian Splatting and World Models — WesEklund · 2026-09-15
- Sidewalk Robots Already Navigate Real Urban Intersections — CurieuxExplorer · 2026-09-15
- GPT-6 Astra gets a Unitree humanoid into a Tesla Model 3 in MuJoCo simulation — YuXiang_IRVL · 2026-09-15
- Tesla Cybercab Heads to Mainland China for First Public Display — XFreeze · 2026-09-15
- Pi robot runs fully autonomous for hours at a chocolate factory — stacking boxes was the hard part — chelseabfinn · 2026-09-15
- Peking University Releases MobileVLA-R1 2.0 with RL-Enhanced Robot Control — PekingUniversity · 2026-09-15