←
Reward AI85 posts
Reward AI avatar

Reward AI

@RewardAI_

Frontier Intelligence for Any Robot

◷ Joined September 2026▧ 12 Photos and videos
Reward AI avatar
Reward AI@RewardAI_·Sep 14

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19
Reward AI avatar
⟳ Reward AI retweeted
Reward AI@RewardAI_·Oct 27

We stress test OM-1 on last-mile delivery for packages from Amazon, Walmart, DHL, etc in different positions and orientations. Our model can reactively adjust, and let humanoids climb, navigate together with whole-body approaching, and hold packages firmly while climbing down.

10:06
Reward AI avatar
⟳ Reward AI retweeted
Reward AI@RewardAI_·Oct 27

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
CyberRobo avatar
⟳ Reward AI retweeted
CyberRobo@CyberRobooo·Oct 28

Beating Amazon to the punch on last-mile delivery,haha Humanoid delivery workers will soon be on the job. a humanoid robot hops onto a Cybertruck, grabs a package, and places it at the doorstep. good demo --- This also explains why humanoid robots are needed.

0:18

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
Rohan Paul avatar
⟳ Reward AI retweeted
Rohan Paul@rohanpaul_ai·Oct 27

imo, the the impressive part here is grip stability during locomotion, since every step down jolts the load in the hands. also stairs are turning into one of the most useful test benches in humanoid robotics. Their OM-1 model here learns from people wearing a sensorized glove that records hand pose, touch and images, with no teleoperation or on-robot demonstrations.

10:06

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
Yanjie Ze avatar
⟳ Reward AI retweeted
Yanjie Ze@ZeYanjie·Oct 27

This video exactly shows why we need bipedal humanoid robots (physical labor in very unstructured envs)

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
Chen Wang avatar
⟳ Reward AI retweeted
Chen Wang@chenwang_j·Oct 27

Excited to show OM-1 taking on last-mile delivery! From picking up packages to carrying them down, it coordinates the whole body and adapts on the fly. We’re pushing into more tasks and environments behind the scenes. More to come!

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
Wenlong Huang @ CoRL 2026 avatar
⟳ Reward AI retweeted
Wenlong Huang @ CoRL 2026@wenlong_huang·Oct 27

Very impressive! The reactiveness in package handling shows the importance of pushing tracking to the limit in data collection to capture the subconscious human behaviors!

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
Zipeng Fu avatar
⟳ Reward AI retweeted
Zipeng Fu@zipengfu·Oct 27

We deployed OM-1 across an expanding range of humanoid tasks, both indoors and outdoors. From last-mile delivery out of a Cybertruck to manipulation on stairs, OM-1 enables humanoids to reliably tackle useful real-world tasks that other models or form factors struggle with.

We deploy OM-1 on more humanoid tasks in both indoor and outdoor environments. OM-1 can power humanoids to do useful things robustly that other models or form factors cannot. - last-mile delivery from a Cybertruck - manipulation on stairs

1:17
Reward AI avatar
Reward AI@RewardAI_·Oct 27

Our website: rewardai.com/ More details about OM-1 can be found in the following post:

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19
Reward AI avatar
Reward AI@RewardAI_·Oct 27

OM-1 can power humanoids to grasp different objects on stairs from different positions and orientations with feet standing on the same or different stair treads.

0:12
Mengyue Yang avatar
⟳ Reward AI retweeted
Mengyue Yang@Mengyue_Yang_·Sep 15

Efficiency will be the next trending?

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19
Ilir Aliu avatar
⟳ Reward AI retweeted
Ilir Aliu@IlirAliu_·Sep 15

.@RewardAI_ came out of stealth yesterday with OM-1, a robot foundation model trained entirely from human demonstrations. The part I find interesting is how they collect the data. Instead of teleoperating a robot, a person wears a device called the Omnibody Hand and simply does the task themselves. Pick something up, fold laundry, mix a drink, package a phone, or plug and unplug cables. The system records the human hand motion, tactile information, proximity, vision and position. That data can then be used to train OM-1 to control different robot bodies. Reward calls the idea: “One Model, One Data Interface, Any Body.” According to the company, a new task can be learned from less than 30 minutes of human demonstration data, without collecting additional robot data. The videos are worth watching. They show four robot arms packaging a phone, robots folding laundry and mixing cocktails, an arm unplugging an Ethernet cable, and early experiments with humanoids. Some of the longer tasks finish in under 30 seconds, in real time. There is also some history behind this. Reward AI builds on DexCap, work that came out of Stanford in 2024 around capturing human dexterous manipulation and transferring it to robots. @chenwang_j , C. Karen Liu and Li Fei-Fei were among the researchers involved. I think the bigger bet here is about the data layer. Robot hardware is changing quickly. If collecting training data means teleoperating every new robot body for thousands of hours, scaling gets expensive fast. Reward is betting that you can collect much of that knowledge from humans instead, then transfer it across different embodiments. That could become quite valuable if it works. For now, we only have the launch material. There is no paper or open model yet, and Reward has not published the kind of large-scale success-rate data needed to judge robustness. Still, this is one of the more interesting approaches I’ve seen to the robotics data problem recently. Source: rewardai.com/blog/OM-1/ —— Weekly robotics and AI insights. Subscribe free: 22astronauts.com

0:17

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19
CyberRobo avatar
⟳ Reward AI retweeted
CyberRobo@CyberRobooo·Sep 15

Honestly, watching robot bartenders, robot installers, and humanoid assistants actually get work done autonomously is way more interesting than sitting around worrying about the dangers of AI. They’re already moving fast and smoothly. Desktop manipulation tasks seem almost effortless, and they look pretty close to being ready for fast customer order fulfillment.

0:30

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19
Raphael cohen avatar
⟳ Reward AI retweeted
Raphael cohen@cohenrap·Sep 15

Why isn't laundry folding the first frame? 😍😍

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19
Afiz ⚡️ avatar
⟳ Reward AI retweeted
Afiz ⚡️@itsafiz·Sep 15

Turning a nut calls for a different kind of control than shaking a cocktail. The range from quick arm movements to careful finger work is pretty compelling here.

Introducing OM-1, our first robot foundation model, zero-shot generalizing to any robot: table-top arms, industrial arms and humanoids. - learned directly from human manipulation data - no teleop/robot data - close to human-level dexterity and efficiency - multi-robot collab

2:19