The Institute of Foundation Models Unveils PAN 2, an Open World Model for Physical Intelligence

Sunnyvale, California, November 10, 2026 — The Institute of Foundation Models (IFM), established by Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), today released PAN 2, an open, general-purpose world model that brings physical intelligence within reach. Give it a starting scene and an action in words, and PAN 2 simulates a video of what happens next, grounded in physics, time, and space.

A language model predicts the next word. A world model simulates the next world, letting people and AI agents ask, “what happens if I do this?” and see the answer before they act. IFM researchers call this simulative reasoning: a way for agents to learn and plan without costly, slow, or irreversible real-world trials.

“A language model learns the world as it has been written down. That is book knowledge, and it takes you a long way, but you cannot learn to swim from a book,” said Eric Xing, Founder of IFM and President and University Professor of Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). “Physical intelligence begins when a system can simulate what its actions would do. PAN 2 is our foundation for that.”

Two kinds of reasoning, one powerful system

• Conceptual reasoning: A large language model works out what will happen and when, producing a time-stamped description of events, actions, and camera movement.
• Perceptual reasoning: A video diffusion model, trained from scratch by IFM on real-world videos chosen for physical dynamics rather than visual appeal, turns that description into video, filling in details like how a wave breaks or how a hand grips a pipette.

PAN 2 builds on the Generative Latent Prediction (GLP) architecture pioneered by Xing and IFM researchers, which predicts change in an internal representation and then reconstructs what would be observed, so predictions can be checked against the real world.

Standout results

• Six core dimensions (physical consistency, causal coherence, entity persistence, constraint satisfaction, viewpoint invariance and objective dynamics): PAN 2 placed just behind MiniMax-H3 and ahead of four other models.
• Multi-step planning: PAN 2 completed more full plans than any other model tested.
• Public benchmarks: Third overall on the Physics-IQ Verified leaderboard and first among open-weight models.

PAN 2 can also simulate situations absent from its training data, such as an egg dropped onto a wide range of surfaces, and can serve as a world where AI agents explore and learn, like cooking a pot of pasta in a kitchen they’ve never seen. Long-term world consistency remains an open challenge, detailed in the technical blog.

Open for everyone to build on

IFM is releasing PAN 2 with video model weights, full training and inference code, a technical blog, and documentation, all under the Apache 2.0 license. Few world models are released with their full training code.

“We want researchers to be able to test PAN 2’s capabilities, understand its limits, and build on it,” said Hector Liu, director of IFM’s Silicon Valley lab. “That means releasing the model weights alongside the training and inferencing code.”

PAN 2 is available now through IFM’s website, ready to be adapted for fields from robotics to education.

About the Institute of Foundation Models

IFM is a global AI research lab dedicated to open, independent development of frontier-class foundation models. Launched in May 2025 by MBZUAI, it operates labs in Abu Dhabi, Silicon Valley, and Paris. Its portfolio includes the K2 series of fully open-source LLMs, Jais, the most advanced Arabic LLM, and PAN, the next-generation world model for physical intelligence.

About Mohamed bin Zayed University of Artificial Intelligence

MBZUAI is a research university in Abu Dhabi, founded in 2019, it enrolled its first graduate students in 2021 and opened a highly selective undergraduate program in 2025. Recognized globally for its research excellence, it has nearly a thousand students from more than 70 countries. Learn more at mbzuai.ac.ae.