A world model is an AI system that learns a predictive representation of how an environment changes over time. It can use observations, actions, or language to estimate future states and possible outcomes before an agent acts. World models support planning in model-based reinforcement learning, robotics, embodied AI, autonomous driving, and interactive simulation. This site catalogs these systems with model records, research papers, benchmark context, and source links. Start with the world models database, compare systems on the leaderboard, or read the Performance Index methodology.
Download the current CC0 export of the world-models.io database, including model metadata, benchmark records, source links, and a validation schema.
This research dataset covers model architecture, benchmark evidence, training method, modality, source links, and research citations. Records are curated and updated as the underlying research changes.
Release: 63 model records · version 1.1.0 · updated 2026-09-01 · CC0 1.0
Canonical Dataset repository on GitHub · JSON · CSV · JSON Schema
Browse the database · View the leaderboard · Read the methodology
A practical interpretation framework for readers comparing fast-moving world-model research.
Scope. AI World Models Dataset is evaluated as a focused research and discovery surface, not as a claim that every listed system is directly comparable. Architecture, task, observation modality, action conditioning, training data, and evaluation protocol determine which comparisons are meaningful.
Evidence method. Prefer original papers, official technical reports, project pages, and versioned benchmark protocols. A reported score is retained with its source and context; it should only influence a ranking when the task, metric, unit, split, and protocol are compatible.
Limitations. World-model terminology changes across reinforcement learning, video generation, robotics, autonomous driving, and embodied AI. Use the linked model and methodology pages to check freshness, reproducibility, availability, and domain fit before treating an editorial summary as a deployment recommendation.