01 / Open source
PyTorch World
A shared world for world-model research.
Open sourcePython / PyTorchApache 2.0
Intelligence,
beyond the visible.
01 / Introduction
Tools for intelligence that understands the world.
02Our first horizon
Selected project / 001The ambitionA world worth
A world worth
understanding.
Better tools for the next generation of world models.
Areas of application
Where we see world models making a difference.
- 01
Advanced Robotics
Helping robots understand their surroundings, anticipate change, and plan actions in the physical world.
- 02
Autonomous Agent Verification
Exploring agent behaviour in simulated scenarios to evaluate decisions before they reach the real world.
- 03
Enterprise & Complex System Simulation
Modelling connected systems to explore scenarios, understand dependencies, and inform decisions.
03Beyond the pixels
Our field of viewRepresentation → Prediction → Actionz → z′
Understanding what is.
Imagining what comes next.
World-model researchLearn the structure.
Learn the structure.
Explore the future.
World models learn how an environment changes. By predicting in a learned representation, they can imagine outcomes and help plan the next action.
Our first toolkit brings together joint-embedding architectures, action-conditioned models, and the environments to explore them.
Explore our researchBuild with us.
Follow the work. Bring your ideas.
Find us on GitHub