1X World Model
$ 0
Physics-based world model using video/prompts; enables NEO robots to learn by observation
Out of stock
Brain Score
2
Specifications and details:
| Nationality | Norway |
|---|---|
| Website | https://www.1x.tech/ |
| Model type | Foundation Model |
| Manufacturer | Unitree Robotics |
| Release date | 2025/2026 |
Description
1X World Model enables humanoid robots to understand and predict physical dynamics by observing and interpreting visual data. Instead of relying on preset routines, it uses video and contextual cues to learn how objects move and interact. As a result, robots adjust their actions more naturally and effectively. Moreover, the model anticipates outcomes before acting, improving reliability and efficiency. Therefore, humanoids can respond smoothly to changing conditions and unexpected situations.
2026 Humanoid Robot Market Report
160 pages of exclusive insight from global robotics experts – uncover funding trends, technology challenges, leading manufacturers, supply chain shifts, and surveys and forecasts on future humanoid applications.
Featuring insights from
Aaron Saunders, Former CTO of
Boston Dynamics,
now Google DeepMind
2026 Humanoid Robot Market Report
160 pages of exclusive insight from global robotics experts – uncover funding trends, technology challenges, leading manufacturers, supply chain shifts, and surveys and forecasts on future humanoid applications.
In addition, 1X World Model integrates perception with physical reasoning to support better decision-making in complex environments. Consequently, robots can plan upcoming steps based on observed patterns. Its design emphasizes learning from real-world interactions rather than only simulations, which enhances performance over time. By focusing on practical observation, the system becomes more capable with each task. Over time, the model aims to increase adaptability and robustness for a wide range of general-purpose humanoid tasks.
Contact Humanoid.guide
Website: https://www.1x.tech
