--- title: "Google DeepMind | World Models Research | world-models.io" canonical: "https://world-models.io/en/labs/deepmind/" description: "Google DeepMind's contributions to AI world models including DreamerV3, MuZero, Genie 2, and the RSSM architecture." author: "Bernard Grenat" generated: "2026-10-30" --- New: the Timeline is live. Track world model releases, papers, and benchmark updates in real time. The Knowledge Hub for AI World Models # Google DeepMind Google DeepMind is the most prolific contributor to world model research, creating DreamerV3, MuZero, Genie 2, the RSSM architecture, and foundational work on model-based RL. robotics model-based-rl simulation embodied-ai ## What is a world model? 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](https://world-models.io/en/world-models-database/), compare systems on the [leaderboard](https://world-models.io/en/leaderboard/), or read the [Performance Index methodology](https://world-models.io/en/methodology/performance-index/). ## About DeepMind Main editorial body preserved directly in static HTML. Google DeepMind is one of the world's leading AI research labs, pioneering work in reinforcement learning, world models, and foundation models. Their contributions include the Dreamer family (PlaNet, DreamerV2, DreamerV3), MuZero, the Predictron, Genie 2, and foundational work on model-based RL. DeepMind's world model research spans from abstract planning models to large-scale generative environment simulators, making it the single most prolific contributor to the field. Google DeepMind is a industry organization based in United Kingdom / United States with a visible footprint in AI world models. ## Lab Snapshot Static summary of the lab profile for crawlers and no-JS readers. | Attribute | Value | | --- | --- | | Lab | Google DeepMind | | Short Name | DeepMind | | Type | industry | | Region | United Kingdom / United States | | Related Models | 13 | | Description | Google DeepMind is one of the world's leading AI research labs, pioneering work in reinforcement learning, world models, and foundation models. Their contributions include the Dreamer family (PlaNet, DreamerV2, DreamerV3), MuZero, the Predictron, Genie 2, and foundational work on model-based RL. DeepMind's world model research spans from abstract planning models to large-scale generative environment simulators, making it the single most prolific contributor to the field. | ## Focus Areas Focus areas surfaced directly from the lab record. model-based-rl generative-world-models planning games robotics ## Related Models Models currently associated with this lab in the local knowledge base. | Model | Category | Year | Index v1.1 | | --- | --- | --- | --- | | [DreamerV3](https://world-models.io/en/models/dreamer-v3/) | Model-Based RL | 2023 | 88/100 | | [DreamerV2](https://world-models.io/en/models/dreamer-v2/) | Model-Based RL | 2021 | 72/100 | | [PlaNet](https://world-models.io/en/models/planet/) | Model-Based RL | 2019 | 57/100 | | [RSSM](https://world-models.io/en/models/rssm/) | Latent Dynamics | 2019 | 64/100 | | [MuZero](https://world-models.io/en/models/muzero/) | Model-Based RL | 2020 | 78/100 | | [Predictron](https://world-models.io/en/models/predictron/) | Model-Based RL | 2017 | 43/100 | | [Genie 2](https://world-models.io/en/models/genie-2/) | Generative World Model | 2024 | 79/100 | | [Genie](https://world-models.io/en/models/genie-1/) | Generative World Model | 2024 | 57/100 | | [UniSim](https://world-models.io/en/models/unisim/) | Generative World Model | 2023 | 72/100 | | [Imagination-Augmented Agents (I2A)](https://world-models.io/en/models/imagination-augmented-agents/) | Model-Based RL | 2017 | 45/100 | | [World Models (Ha & Schmidhuber)](https://world-models.io/en/models/ha-world-model/) | Model-Based RL | 2018 | 48/100 | | [Pandora](https://world-models.io/en/models/pandora/) | Generative World Model | 2024 | 52/100 | | [RT-2](https://world-models.io/en/models/rt-2/) | Foundation World Model | 2023 | 72/100 | ## Frequently Asked Questions FAQ answers rendered directly into static HTML for extractable responses. ### What world models has Google DeepMind created? DeepMind has created PlaNet, the RSSM architecture, DreamerV1/V2/V3, MuZero, the Predictron, Genie 1 & 2, Imagination-Augmented Agents, and UniSim. They are the most prolific contributor to world model research. ### What is DeepMind's most important world model? DreamerV3 is widely considered their most impactful world model, achieving the first domain-general agent that masters diverse environments with fixed hyperparameters. ## Quick Answer Short extractable summary preserved directly in static HTML. - Google DeepMind is a industry organization based in United Kingdom / United States with a visible footprint in AI world models. - On world-models.io, it is associated with 13 world models across editorially linked pages. - Use this page to review the organization profile, connected models, related comparisons, and official sources. ## Editorial Trust Signals Editorial provenance and refresh policy preserved directly in static HTML. **Published by** world-models.io editorial board. **Lead editor** Bernard Grenat. This lab page distinguishes official organization sources from world-models.io editorial synthesis and related model coverage. Each editorial page is assembled from primary sources, normalized into extractable summaries, checked for factual drift, and reviewed before publication or major refreshes. Last reviewed: 2026-09-27. Pages are refreshed when a new paper, benchmark, release, architecture update, or stronger primary source materially changes the answer a reader or AI system should retrieve. Each page links back to relevant primary sources and keeps a stable canonical URL so readers can verify claims, trace context, and reference the most up-to-date version. [See the editorial policy.](https://world-models.io/en/editorial-policy/) Primary sources only Last reviewed date visible Methodology documented Source links included Correction channel published Review date visible ## Official Sources Primary lab and model sources embedded in static HTML. - [Google DeepMind official website](https://deepmind.google) Primary organizational source - [DreamerV3 paper](https://arxiv.org/abs/2301.04104) Google DeepMind primary research source - [DreamerV2 paper](https://arxiv.org/abs/2010.02193) Google primary research source - [DreamerV2 cited source](https://arxiv.org/abs/1803.10122) Hafner et al., 2021. Mastering Atari with Discrete World Models. ICLR 2021.