1X TechnologiesUnited States2026
1X World Model (1XWM)
An AI model: a vision-language-action model (VLA) takes camera images and an instruction and turns them into movements for a robot. This page describes the software, not a robot.
Key figures as stated by the developer
2 of 6 stated · 4 gaps- Model type not stated
- Controls not stated
-
Inputs A
cameralanguage - Access not stated
- Parameters not stated
- Base model A a 14B generative video model (not named)
How to read the figures
- 14kgA30.9lbA
- The letter in the grey square after a figure points to the source the figure came from. Robots with several sources carry several letters.
- 74.5lb
- Imperial figures are our own conversion unless the manufacturer states an imperial figure itself. The source letter always belongs to the metric figure — a conversion has no source of its own.
Versions and notes in the key figures (5)
Model type
Note The developer does not label it vision-language-action, world action model or embodied reasoning: "we introduce our video-pretrained world model, 1XWM, integrated into NEO as a robot policy.".
Controls
Note The model generates video, and a separate inverse dynamics model extracts the robot actions: "The WM rolls out the intended future, the IDM extracts the necessary trajectory, and the robot executes the sequence in the real world."
Inputs
Blog: "At inference time, the system receives a text prompt and a starting frame."
Parameters
Note The base of the backbone: "The 1XWM backbone is built upon a 14B generative video model."
Base model
Blog: "The 1XWM backbone is built upon a 14B generative video model."
Full specifications
3 of 12 fields stated| Field | Value | Source and caveats |
|---|---|---|
| Input and output | ||
| Model type | not stated | Note The developer does not label it vision-language-action, world action model or embodied reasoning: "we introduce our video-pretrained world model, 1XWM, integrated into NEO as a robot policy.". A |
| Inputs | camera language | A Blog: "At inference time, the system receives a text prompt and a starting frame." |
| Controls | not stated | Note The model generates video, and a separate inverse dynamics model extracts the robot actions: "The WM rolls out the intended future, the IDM extracts the necessary trajectory, and the robot executes the sequence in the real world." A |
| Degrees of freedom controlled | not stated | |
| Size and speed | ||
| Parameters | not stated | Note The base of the backbone: "The 1XWM backbone is built upon a 14B generative video model." A |
| Control rate | not stated | Note Latency, not a control rate: "the 1XWM backbone takes 11 seconds to run, using multi-gpu inference". "Each inference generates 5 seconds of real-time video, and the IDM takes 1 second to extract actions from the generated rollout." A |
| Where it runs and what it is built on | ||
| Runs where | not stated | Note Blog: "we send a prompt from our frontend to the model inference server, which will then execute the action on-robot." A |
| Base model | a 14B generative video model (not named) | A Blog: "The 1XWM backbone is built upon a 14B generative video model." |
| Openness | ||
| Access | not stated | |
| License | not stated | |
| The developer’s own claims | ||
| Learns new tasks | without examples | A Developer claim: "our world model generalizes to novel objects, motions, and tasks without pre-training on large-scale robot data or any related teleoperated demonstrations". "At inference time, the system receives a text prompt and a starting frame." |
| Works across robots | not stated | 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. |
Notes
- 1X Technologies describes 1XWM as a video-pretrained world model that is integrated into NEO as a robot policy: it generates video of the intended future from a text prompt and a starting frame, and a separate model turns that video into robot actions.
- The name 1X World Model is older: 1X first showed a world model that works as a virtual simulator for its robots on 17 September 2024, and described the 1XWM evaluation use on 16 June 2025; the policy use was published on 12 January 2026.
Sources
-
A
Manufacturer product page. www.
1x. · Retrievedtech/ discover/ world-model-self-learning
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.