NVIDIA Launches Rubin Platform and Physical AI Models at CES 2026

NVIDIA CEO Jensen Huang unveils Rubin architecture and open physical AI models at CES 2026, positioning company as Android of robotics with new humanoid robot capabilities and partnerships.

by Cody RodeoUpdated Feb 17, 2026 • 6:07 PM
NVIDIA Launches Rubin Platform and Physical AI Models at CES 2026Photo by Alex Knight on Unsplash

NVIDIA CEO Jensen Huang opened CES 2026 in Las Vegas with the announcement of Rubin, the company's first extreme-codesigned AI platform designed to replace the Blackwell architecture starting in the second half of 2026. The unveiling positions NVIDIA to become the default platform for generalist robotics, mirroring Android's dominance in smartphones.

Related / CES 2026 Highlights Stablecoin Payment Revolution and Digital Wallet Growth

Physical AI emerged as CES 2026's defining theme, taking center stage from last year's agentic AI focus. NVIDIA defines physical AI as models trained in virtual environments using synthetic computer-generated data, then deployed as physical machines once they master their intended purpose. This approach dramatically reduces training costs while enabling rapid iteration on robot behaviors.

The company released several groundbreaking open AI models including NVIDIA Cosmos Transfer 2.5 and Cosmos Predict 2.5 for world modeling, Cosmos Reason 2 vision language model that enables machines to see and understand the physical world, and Isaac GR00T N1.6 specifically designed for humanoid robot full-body control. NVIDIA also unveiled the Alpamayo family of models for autonomous vehicles.

Global industry leaders including Boston Dynamics, Caterpillar, Franka Robotics, Humanoid, LG Electronics, and NEURA Robotics are leveraging NVIDIA's robotics stack to debut AI-driven robots. The Blackwell architecture-powered Jetson T4000 module is now available, delivering 4x greater energy efficiency and AI compute for edge deployment.

Industry analysts suggest NVIDIA's comprehensive robotics ecosystem—spanning simulation, training, and deployment—positions the company to capture significant value as physical AI transitions from laboratory demonstrations to mainstream commercial applications across manufacturing, logistics, healthcare, and consumer markets throughout 2026 and beyond.