Reward AI's OM-1 Learns Robot Manipulation from Humans
Loading video
Loading videoReward AI's OM-1 learns manipulation from wearable human demonstrations, with no teleoperation or on-robot training data, across robot arms and humanoids.
Embodied Foundation ModelHuman ManipulationRobot ManipulationDexterous ManipulationCross-EmbodimentReinforcement LearningHumanoid robotsRobot ArmsReward AIOmnibody HandOM-1Embodied Base ModelHumanoid
Category: humanoid
Author: @humanoidsdaily
Date: 2026-09-14T00:00:00
Duration: 36.564s
Reference: https://rewardai.com/blog/OM-1/
