RewardAI unveils OM-1 robot foundation model with zero-shot generalization across robots

zipengfu · x · 2026-09-15

RewardAI introduced OM-1, its first robot foundation model that zero-shot generalizes to any robot embodiment: tabletop arms, industrial arms, and humanoids.

Researcher animeshgarg shared the demo, praising its speed and overall generality; the work comes from zipengfu, chenwangj and team. If the claims hold, replacing teleop data collection with human manipulation data marks a notable shift in robot training paradigms.

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

Original post →

More from Embodied

Embodied channel →