Reward AIUnited States2026
OM-1
An AI model: software that a robot uses to perceive, plan or move. This page describes the software, not a robot.
Robotsnot stated
rewardai.com
Key figures as stated by the developer
2 of 6 stated · 4 gaps- Model type not stated
-
Controls A
arms and hands -
Inputs A
cameratouch - Access not stated
- Parameters not stated
- Base model not stated
How to read the figures
- 50HzA
- The letter in the grey square after a figure points to the source the figure came from. Pages with several sources carry several letters.
Versions and notes in the key figures (3)
Model type
Note Reward AI's own term is "our in-house general-purpose robot policy and the decision-making core of the system".
Controls
Blog post: "The actions OM-1 outputs carry motion direction, speed, force, and the timing of key events such as grasping and moving." "We start with the hand as a shared interface for learning manipulation across these bodies."
Inputs
Blog post: "OM-1 takes as input the multimodal streams collected by Omnibody Hand: images, tactile signals, inter-finger proximity, and hand pose trajectories." Images = camera, tactile signals = touch.
Full specifications
3 of 13 fields stated| Field | Value | Source and caveats |
|---|---|---|
| Input and output | ||
| Model type | not stated | Note Reward AI's own term is "our in-house general-purpose robot policy and the decision-making core of the system". A |
| Inputs | camera touch | A Blog post: "OM-1 takes as input the multimodal streams collected by Omnibody Hand: images, tactile signals, inter-finger proximity, and hand pose trajectories." Images = camera, tactile signals = touch. |
| Controls | arms and hands | A Blog post: "The actions OM-1 outputs carry motion direction, speed, force, and the timing of key events such as grasping and moving." "We start with the hand as a shared interface for learning manipulation across these bodies." |
| Size and speed | ||
| Control rate | not stated | Note Blog post, control layer: "running underneath at high frequency". A |
| The developer’s own claims | ||
| Learns new tasks | not stated | Note Developer claim, in terms of human demonstration data: OM-1 "picks up a brand-new task, including challenging dynamics and long horizons, from less than 30 minutes of data". A |
| Works across robots | yes | A Developer claim: the policy "runs across robot bodies, from industrial arms to humanoids". Yes only if the developer itself says the model runs on different robot models, for example robots from two manufacturers or two of one manufacturer’s models. Several units of the same robot do not count. |
Not stated: Degrees of freedom controlled, Parameters, On robot or cloud, Base model, Access, License, Free or paid
Notes
- Reward AI calls OM-1 its in-house general-purpose robot policy and the decision-making core of its Omnibody system.
- Reward AI published OM-1 on its blog in September 2026.
- Reward AI's X profile gives its location as Bay Area, CA.
Sources
-
A
Primary source. www.
rewardai. · Retrievedcom/ blog/ OM-1/
One line per source, not per figure: a source has one retrieval date. The letter beside a figure points to the line it came from.