Artificial IntelligenceThursday, November 5, 20261 min read

New NVIDIA Cosmos 3 Open Models for Interactive World Simulation

By Engineering Blogs · Published by Everything Blog

In short

The release introduces four distinct checkpoints built as 16-billion-parameter variants of Cosmos 3 Nano under the OpenMDW1.1 license. Developers can utilize these tools to evaluate driving policies across diverse environmental conditions or model complex physical behaviors from direct control inputs. Notably, Cosmos3-Nano-Sim-Auto enables efficient ongoing simulation by producing synchronized camera and LiDAR streams in only a few denoising steps.

Key points

Developers can deploy open-weight world models to generate synthetic multi-sensor driving…: Developers can deploy open-weight world models to generate synthetic multi-sensor driving data across varied weather and lighting conditions.
The causally distilled simulation models enable fast streaming of synchronized sensor fee…: The causally distilled simulation models enable fast streaming of synchronized sensor feeds using only a few denoising steps.
Robotics researchers gain tools to convert physics simulator controls and text prompts in…: Robotics researchers gain tools to convert physics simulator controls and text prompts into photorealistic video observations.

Sources checked by our analysis pipeline: 1

TL;DR

NVIDIA Cosmos world model development platform adds four new open-weight checkpoints to its Cosmos 3 model family for generating interactive world simulation and multi-sensor multi-view driving scenes:

- 🌦️ Cosmos3-Nano-Transfer-Auto : Generate multi-view driving data with varied weather, lighting, and time of day.

- 🚗 Cosmos3-Nano-Sim-Auto : Stream synchronized camera and LiDAR observations for driving simulation.

- 🎨 Cosmos3-Nano-Sim-Transfer : Turn simulator controls and a text prompt into photorealistic video.

- 🤖 Cosmos3-Nano-Sim-Bimanual : Predict video from robot or camera actions.

All four models are post-trained examples of Cosmos 3 Nano and have 16B trainable parameters. The models use OpenMDW1.1, permitting commercial and non-commercial use under its terms.

Download the checkpoints · Read the technical report . Watch the demos

Why do you need these checkpoints?

🦾 Robotics needs environments that can answer this across changing scenes and viewpoints. These checkpoints offer complementary ways to build them: Sim-Bimanual predicts visual consequences from actions, while Sim-Transfer turns a physics simulator’s controls into generated observations.

🚦 For autonomous driving, appearance and geometry must agree. Synchronized camera and LiDAR generation connects visual…

AI Analysis

Tone

Positive

In short

Key Points

Developers can deploy open-weight world models to generate synthetic multi-sensor driving…: Developers can deploy open-weight world models to generate synthetic multi-sensor driving data across varied weather and lighting conditions.
The causally distilled simulation models enable fast streaming of synchronized sensor fee…: The causally distilled simulation models enable fast streaming of synchronized sensor feeds using only a few denoising steps.
Robotics researchers gain tools to convert physics simulator controls and text prompts in…: Robotics researchers gain tools to convert physics simulator controls and text prompts into photorealistic video observations.

Topics

Artificial Intelligence

Was this summary useful?

AI Analysis

Tone

Positive

In short

Key Points

Developers can deploy open-weight world models to generate synthetic multi-sensor driving…: Developers can deploy open-weight world models to generate synthetic multi-sensor driving data across varied weather and lighting conditions.
The causally distilled simulation models enable fast streaming of synchronized sensor fee…: The causally distilled simulation models enable fast streaming of synchronized sensor feeds using only a few denoising steps.
Robotics researchers gain tools to convert physics simulator controls and text prompts in…: Robotics researchers gain tools to convert physics simulator controls and text prompts into photorealistic video observations.

Topics

Artificial Intelligence

Was this summary useful?

METR: Anthropic redacted much of the "most informative" evidence from METR's 100-page report, so readers "must rely on our judgment more than we consider ideal" (@metr_evals)
Independent evaluation group METR publicly disclosed that substantial portions of critical findings were withheld from its final assessment. These extensive omissions limit external researchers and the public from independently verifying the underlying data. Consequently, outside observers are forced to depend heavily on the organization's unsupported conclusions rather than raw evidence.
1 min read
Modernize your .NET apps for Windows Arm64 with GitHub Copilot
Transitioning software to power-efficient infrastructure often introduces complex compatibility hurdles, particularly when dealing with legacy architecture settings and unsupported native binaries. Developers can now utilize an automated assistant to evaluate their codebases, resolve blockers, and generate customized transition roadmaps. This support aims to accelerate transitions toward cheaper compute environments without requiring manual rewrites of core project configurations.
2 min read
Getting started with Web Answer on Amazon Bedrock AgentCore
The managed tool allows AI developers to instantly query external websites and receive fully synthesized, attributed answers without maintaining bespoke retrieval architectures. Operating via an AgentCore Gateway, the feature executes multi-step web searches, cross-references sources, and appends citations directly to natural-language outputs. This turnkey solution eliminates the infrastructure overhead of managing raw web scraping and synthesis pipelines.
8 min read

New NVIDIA Cosmos 3 Open Models for Interactive World Simulation

In short

Key points

More in Artificial Intelligence