Editorial definition preserved directly in static HTML.
Foundation world models are large-scale, general-purpose models trained on massive datasets to learn broad representations of world dynamics. They aim to serve as versatile base models for diverse downstream tasks, from robotics to autonomous driving to video generation.
Large-scale world models designed as general-purpose foundations for diverse embodied AI tasks.
Static category overview generated from local editorial data.
| Attribute | Value |
|---|---|
| Category | Foundation World Models |
| Description | Large-scale world models designed as general-purpose foundations for diverse embodied AI tasks. |
| Definition | Foundation world models are large-scale, general-purpose models trained on massive datasets to learn broad representations of world dynamics. They aim to serve as versatile base models for diverse downstream tasks, from robotics to autonomous driving to video generation. |
| Related Models | 4 |
| Related Research | 1 |
| Related Guides | 0 |
Additional editorial context preserved directly in static HTML.
Foundation world models aim to provide reusable physical priors at scale, reducing the need to rebuild every simulator or planning backbone from scratch for robotics, driving, or other embodied tasks.
The real question is not only parameter count or dataset size. It is whether the model transfers across downstream settings, exposes meaningful evidence of generality, and supports practical adaptation in physical AI workflows.
| Model | Lab | Category | Year |
|---|---|---|---|
| NVIDIA Cosmos | NVIDIA | Foundation World Model | 2024 |
| Genie 2 | Google DeepMind | Generative World Model | 2024 |
| GAIA-1 | Wayve | Foundation World Model | 2023 |
| AMI World Model | AMI Labs | Foundation World Model | 2024 |
| Topic | Summary |
|---|---|
| Foundation World Models | How foundation world models such as Cosmos and Genie 2 bring large-scale learned simulation to robotics, autonomous driving, and physical AI. |
FAQ answers rendered directly into static HTML for extractable responses.
Foundation world models are large-scale reusable models trained on broad video or multimodal corpora to support downstream robotics and physical AI tasks.
They can provide reusable world knowledge at scale, reducing the need to train every simulator or planning model from scratch.
Short extractable summary preserved directly in static HTML.
Editorial provenance and refresh policy preserved directly in static HTML.
Published by world-models.io editorial board.
Lead editor Bernard Grenat.
This category page maintains a stable editorial definition and connects it to related models, research topics, guides, and source-backed context.
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.
Representative external references connected to this category through related models and research topics.