Reward AI debuts OM-1 robot foundation model, zero-shot generalizing to any robot

yifengzhu_ut · x · 2026-09-15

Reward AI, founded by Yifeng Zhu (PhD and big-tech lab background in robot manipulation), introduced OM-1, its first robot foundation model. Key claims: zero-shot generalization across table-top arms, industrial arms, and humanoids; learned directly from human manipulation data with no teleop or robot data; near human-level dexterity and efficiency; multi-robot collaboration. The team treats cross-embodiment and efficiency as first-class citizens, aiming to deploy millions of robots of varied form factors in the real world.

Related event: Physical Intelligence Unveils OM-1, a Robot Foundation Model Trained on Human Data(22 posts)→

Original post →

More from Embodied

Embodied channel →