NVIDIA Launches Cosmos 3, the Open Frontier Foundation Model for Physical AI - NVIDIA Newsroom
Quick Answer
NVIDIA has launched Cosmos 3, an open physical AI foundation model that utilizes a mixture-of-transformers architecture for advanced reasoning and simulation.
Quick Take
NVIDIA has launched Cosmos 3, an open foundation model that utilizes a mixture-of-transformers architecture for advanced reasoning and simulation. This model supports multimodal generation across various formats and is backed by a coalition of leading AI and robotics organizations to enhance open world model development.
Key Points
- Cosmos 3 is the first fully open omnimodel for physical AI applications.
- It features native vision reasoning and multimodal generation capabilities.
- The launch includes the NVIDIA Cosmos Coalition with top AI and robotics labs.
- Designed for synthetic data generation and physical AI policy model development.
- Built on a breakthrough mixture-of-transformers architecture.
Article Excerpt
From source RSS / original summary# NVIDIA Launches Cosmos 3, the Open Frontier Foundation Model for . * NVIDIA Cosmos 3 is a new leaderboard-topping open physical AI foundation model, built on a breakthrough mixture-of-transformers architecture for physical AI reasoning, world simulation and action generation.
* Cosmos 3 is the world’s first fully open omnimodel with native vision reasoning and multimodal generation across text, image, video, ambient sound and action for state-of-the-art synthetic data generation and physical AI policy model development. * NVIDIA launches the NVIDIA Cosmos Coalition with leading AI labs and robotics leaders — including Agile Robots, Black Forest Labs, Generalist, LTX, Runway and Skild AI — to advance the next generation of open world models. **NVIDIA GTC Taipei**—NVIDIA today l
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from WebSearch (Tavily)
See more →Stop just chatting with AI. Learn to build production-ready software in ...
The 2026 Bootcamp offers hands-on training in building production-ready software using Generative AI, LLM applications, and AI agents, emphasizing practical skills over casual interaction with AI. Participants will learn to develop applications like Cursor AI, preparing them for real-world challenges in AI development.