RewardAI unveils OM-1 robot foundation model trained from human manipulation data, no teleop

zipengfu · x · 2026-09-15

RewardAI introduced OM-1, its first robot foundation model that zero-shot generalizes to tabletop arms, industrial arms and humanoids. It learns directly from human manipulation data (no teleop or robot data), approaches human-level dexterity and efficiency, and supports multi-robot collaboration. Reposters call the policy speed another level, teasing a bigger announcement.

Related event: Reward AI Unveils OM-1, a Robot Foundation Model Trained Purely on Human Data with Zero-Shot Generalization(17 posts)→

Original post →

More from Embodied

Embodied channel →