Home Original page

NVIDIA Cosmos

Develop world foundation models to advance physical AI.

What Is NVIDIA Cosmos?

Cosmos Cookbook

This cookbook serves as a practical guide to Cosmos open models. It offers step-by-step workflows, technical recipes, and concrete examples for building, adapting, and deploying WFMs.

World Foundation Models for Physical AI

Open and fully customizable pretrained models for world generation and understanding.

Cosmos Predict

Predict future states of dynamic environments for robotics and AI agent planning.

This world generation model produces up to 30 seconds of high-fidelity video from multimodal prompts.

Cosmos Transfer

Accelerate synthetic data generation across various environments and lighting conditions.

This multicontrol model transforms 3D or spatial inputs from physical AI simulation frameworks, such as CARLA or NVIDIA Isaac Sim™, into fully controlled high-fidelity video.

Cosmos Reason

Enable robots and vision AI agents to reason like humans. 

This multimodal vision language model (VLM) leverages prior knowledge, physics understanding, and common sense to comprehend the real world and interact with it.

Data Processing

Speed up efficient dataset processing and generation.

Quickly filter, annotate, and deduplicate large amounts of sensor data necessary for physical AI development with Cosmos Curator. 

You can also instantly query these datasets and retrieve scenarios with NVIDIA Cosmos Dataset Search (CDS).

How Cosmos Accelerates AI Across Industries

Use Cosmos WFMs to simulate, reason, and generate data for downstream pipelines in robotics, autonomous vehicles, and industrial vision systems.

Robot Learning

Robots need vast, diverse training data to effectively perceive and interact with their environments. Cosmos WFMs solve this in multiple ways:

  • Generate synthetic data using Cosmos Transfer.
  • Post-train Cosmos Predict for your robot policy.
  • Reason and filter synthetic data using Cosmos Reason.

Autonomous Vehicle Training

Diverse, high-fidelity sensor data is critical for safely training, testing, and validating autonomous vehicles. But it’s difficult, time-consuming, and costly to scale.

With Cosmos WFMs post-trained on vehicle data, you can:

  • Amplify existing data diversity with new weather, lighting, and geolocation data using Cosmos Transfer.
  • Expand into multi-sensor views using Cosmos Predict.

Video Analytics AI Agents

Enhance automation, safety, and operational efficiency across industrial and urban environments. 

With Cosmos Reason, AI agents can analyze, summarize, and interact with real-time or recorded video streams to:

  • Deliver real-time question-answering and alerts.
  • Provide rich contextual insights.

Get Started With NVIDIA Cosmos

1

Ready to build? Access models and code directly.

2

Not ready to build yet? Try Cosmos models in our hosted catalog.

3

 Need help? Start quickly with our hands-on model recipes.

Supporting the Physical AI Community

Cosmos models, guardrails, and tokenizers are available on Hugging Face and GitHub, with resources to tackle data scarcity in training physical AI models.

Get the Best Performance With NVIDIA Blackwell

NVIDIA RTX PRO 6000 Blackwell Series Servers accelerate physical AI development for robots, autonomous vehicles, and AI agents across training, synthetic data generation, simulation, and inference.

Unlock peak performance for Cosmos world foundation models on NVIDIA Blackwell GB200 for industrial post-training and inference workloads.

Adopted by Leading Physical AI Innovators

Model developers from the robotics, autonomous vehicles, and vision AI industries are using Cosmos to accelerate physical AI development.

Next Steps

Join the Cosmos Community

Connect with Cosmos experts, engage with fellow developers, provide model feedback, and access continued learning through livestreams and recipes.

Cosmos Cookbook

A comprehensive guide for working with the NVIDIA Cosmos ecosystem for real-world, domain-specific applications across robotics, simulation, autonomous systems, and physical scene understanding.

Build Video Analytics AI Agents

Use Cosmos Reason with NVIDIA Blueprint for video search and summarization (VSS) to build AI agents for scalable, real-time video understanding.

The Latest From Cosmos Developers

Frequently Asked Questions