RewardAI unveils OM-1, its first robot foundation model trained purely on human manipulation data

zipengfu · x · 2026-09-15

RewardAI introduced OM-1, its first robot foundation model, zero-shot generalizing to any robot form factor: tabletop arms, industrial arms, and humanoids.

Key claims: it learns directly from human manipulation data with no teleoperation or robot-collected data, achieves near human-level dexterity and efficiency, and supports multi-robot collaboration. Commenters highlighted its speed and smoothness as essential for real-world deployment.

Related event: Physical Intelligence unveils OM-1 robot foundation model(31 posts)→

Original post →

More from Embodied

Embodied channel →