Compare

DEEP Robotics · robot AI model

DR-WM-1.0

China · 2026

Sources retrieved: 27 October 2026 (A, B, C)

2 of 7 key figures stated, as stated by the developer

Model type
Not stated
What it controls
Not stated
Inputs
Camera Robot stateA
Access
Not stated
Parameters
Not stated
Base model
Wan video modelC
Licence type
Not stated

Visit website

Five photos of quadruped robots on a wooded slope, a rocky riverbank, in a stream, on snow and on rubble

Specifications

3/14 specifications disclosed 11 not stated

Not statedThe developer publishes nothing on it

Select a source marker for source details and notes.

Release1/1 disclosed

Released13 October 2026

Input and output1/4 disclosed

Model typeNot stated
InputsCamera Robot state
What it controlsNot stated
Degrees of freedom controlledNot stated

Size and speed0/2 disclosed

ParametersNot stated
Control rateNot stated

Where it runs and what it is built on1/2 disclosed

On robot or cloudNot stated
Base modelWan video model

Openness0/3 disclosed

AccessNot stated
Licence documentNot stated
Licence typeNot stated

The developer’s own claims0/2 disclosed

Learns new tasksNot stated
Works across robotsNot stated

Notes

DEEP Robotics lists DR-WM-1.0 as a world model in its press release of 24 October 2026; its project page and its arXiv preprint of 13 October 2026 call the described model DR-WM.

DEEP Robotics describes DR-WM as predicting how a quadruped's actions play out on different terrain; the rollout takes recorded actions and does not choose them.

DEEP Robotics recorded the Physcene-Robot interaction dataset with its Lynx M20 and Lite3 robots; the model itself takes recorded actions and predicts future motion.

Sources

One line per source, not per figure: a source has one retrieval date.

  1. Aarxiv.org/abs/2610.17630Retrieved 27 October 2026
  2. Bdeveloper.deeprobotics.cn/research/articles/dr-wm/index.htmlMain source. Retrieved 27 October 2026
  3. Carxiv.org/html/2610.17630Retrieved 27 October 2026

Report an error or send an updated datasheet