New: the Timeline is live. Track world model releases, papers, and benchmark updates in real time.
world-models.io
The Knowledge Hub for AI World Models

World Models Database

The most comprehensive database of AI world models, covering model-based RL, generative simulators, foundation world models, and embodied AI architectures.

robotics model-based-rl simulation embodied-ai

Quick Answer

Short extractable summary preserved directly in static HTML.

  • This database answers the question of which world models exist and how to narrow them down by architecture, lab, status, modality, and year.
  • Each record is designed to move a reader from browsing to evaluation with model type, originating lab, benchmark context, and linked research.

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 hub maintains a structured directory of world model records connected to labs, categories, comparisons, and research pages.

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: 2026-06-21.

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.

Primary sources onlyLast reviewed date visibleMethodology documentedSource links included

Database Table

Canonical static snapshot of the world models database. Client-side filters can still enhance this page for interactive users.

ModelLabCategoryStatusModalityYear
HY-World 2.0Tencent HunyuanFoundation World ModelactiveText/Image/Video → Navigable 3D World2026
LeWorldModelMila / NYU / Samsung SAILSelf-Supervised World ModelactiveVisual → Latent Predictions2026
MarbleWorld LabsFoundation World ModelactiveText/Image/Video/360 -> Persistent 3D World2026
PixVerse R1PixVerseGenerative World ModelactiveText/Image → Interactive Shared World2026
PlayWorldPrinceton UniversityGenerative World ModelactiveMulti-view Robot Video + Actions -> Future Observations2026
1X World Model1XFoundation World ModelactiveRobot Video + Actions -> Future Observations2025
GAIA-2WayveGenerative World ModelactiveMulti-View Driving Video + Structured Conditions → Future Driving Video2025
Genie 3Google DeepMindGenerative World ModelactiveText → Interactive 3D Environment2025
Matrix-Game 2.0Skywork AIGenerative World ModelactiveImage + Keyboard/Mouse Actions → Streaming Interactive Video2025
Odyssey-2OdysseyFoundation World ModelactiveText + Video Context → Interactive Video Stream2025
RELICAdobe ResearchGenerative World ModelemergingImage + Text + Camera Actions → Interactive Video2025
V-JEPA 2MetaSelf-Supervised World ModelactiveVideo → Latent Predictions2025
WHAMMicrosoft Research / Ninja TheoryGenerative World ModelactiveGame Video + Controller Actions → Future Video + Actions2025
WHAM-RTMicrosoft Research / Ninja TheoryGenerative World ModelactiveGame State + Actions → Real-Time Interactive Video2025
3D-VLAMIT / TsinghuaFoundation World Modelemerging3D Point Clouds + Language + Actions2024
AMI World ModelAMI LabsFoundation World ModelemergingVision + Language + Proprioception2024
DIAMONDMicrosoft Research / University of GenevaModel-Based RLactiveVisual2024
GameNGenGoogle ResearchGenerative World ModelactiveActions + Frame History → Next Frame2024
Gen-3 AlphaRunwayGenerative World ModelactiveText + Image → Video2024
GenieGoogle DeepMindGenerative World ModelfoundationalImage → Interactive 2D Environment2024
Genie 2Google DeepMindGenerative World ModelactiveImage → Interactive 3D Environment2024
Large World Model (LWM)UC BerkeleyFoundation World ModelactiveVisual (Video) + Language2024
NVIDIA CosmosNVIDIAFoundation World ModelactiveVideo + 3D2024
OASISDecart / EtchedGenerative World ModelactiveActions → Video (real-time)2024
PandoraTsinghua University / ByteDanceGenerative World ModelemergingText + Actions → Video2024
SoraOpenAIGenerative World ModelactiveText → Video2024
TD-MPC2MIT / MetaModel-Based RLactiveState + Visual2024
V-JEPAMetaSelf-Supervised World ModelactiveVideo2024
Copilot4DWaabiFoundation World ModelactiveLiDAR point clouds (4D)2023
DreamerV3Google DeepMindModel-Based RLactiveVisual + Proprioceptive2023
Emu VideoMetaGenerative World ModelactiveText → Image → Video2023
GAIA-1WayveFoundation World ModelactiveVideo + Text + Actions2023
I-JEPAMetaSelf-Supervised World ModelactiveImage2023
IRISMicrosoft ResearchModel-Based RLactiveVisual2023
RT-2Google DeepMindFoundation World ModelactiveVisual + Language + Proprioceptive2023
Stable Video DiffusionStability AIGenerative World ModelactiveVisual (Image → Video)2023
STEVE-1UT AustinGenerative World ModelactiveVisual + Language (instructions)2023
UniSimGoogle DeepMindGenerative World ModelactiveVideo + Actions2023
MILEWayveFoundation World ModelactiveVisual (Multi-camera) + Ego-state2022
Waabi WorldWaabiGenerative World ModelactiveDriving Data + Agent Actions → Reactive Driving Simulation2022
DreamerV2GoogleModel-Based RLfoundationalVisual2021
MuZeroGoogle DeepMindModel-Based RLfoundationalBoard states / Visual (Atari)2020
PlaNetGoogleModel-Based RLfoundationalVisual2019
RSSMGoogleLatent DynamicsactiveVisual + Proprioceptive2019
World Models (Ha & Schmidhuber)Google Brain / IDSIAModel-Based RLfoundationalVisual2018
Imagination-Augmented Agents (I2A)Google DeepMindModel-Based RLfoundationalVisual2017
PredictronGoogle DeepMindModel-Based RLfoundationalAbstract state representations2017
Value Prediction Network (VPN)University of Michigan / Google BrainModel-Based RLfoundationalVisual / Abstract states2017

Featured Models

Representative frontier systems highlighted directly in static HTML.

ModelLabCategorySummary
HY-World 2.0Tencent HunyuanFoundation World ModelTencent Hunyuan's open multimodal world model for reconstructing, generating, and simulating navigable 3D worlds.
LeWorldModelMila / NYU / Samsung SAILSelf-Supervised World ModelA compact 15M-parameter JEPA world model that learns real-world physics on a single GPU, solving the notorious representation collapse problem.
MarbleWorld LabsFoundation World ModelWorld Labs' multimodal world model for generating spatially consistent, persistent 3D environments from text, images, video, and 360 inputs.
PixVerse R1PixVerseGenerative World ModelThe first real-time world model supporting multi-user shared worlds with personalized avatars and no session limits.
PlayWorldPrinceton UniversityGenerative World ModelA Princeton robot world model trained from autonomous self-play to simulate contact-rich manipulation and support policy evaluation and RL fine-tuning.
1X World Model1XFoundation World Model1X's physics-grounded video world model for anticipating the outcomes of NEO's actions and supporting generalization to unseen household tasks.

How To Use This Database

Static usage guidance for non-JS readers.

Use category and lab pages to narrow the field, then open individual model pages for architecture details, benchmarks, citations, and related entities.

For ranked browsing, continue to the leaderboard and the Performance Index methodology. For editorial curation, use the top-world-models and robotics collections.

Frequently Asked Questions

FAQ answers rendered directly into static HTML for extractable responses.

How many world models are in the database?

The static database currently includes 48 world model records sourced from the local knowledge base.

What can I compare from this page?

You can compare model category, originating lab, lifecycle status, modality, and publication year before opening detail pages or side-by-side comparisons.

Quick Answer

Short extractable summary preserved directly in static HTML.

  • This database answers the question of which world models exist and how to narrow them down by architecture, lab, status, modality, and year.
  • Each record is designed to move a reader from browsing to evaluation with model type, originating lab, benchmark context, and linked research.

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 database maintains structured model records connected to labs, categories, comparisons, and research pages for retrieval and verification.

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: 2026-06-21.

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.

Primary sources onlyLast reviewed date visibleMethodology documentedSource links included

External Sources

Primary references and official sources surfaced directly in static HTML for crawlers and no-JS readers.