At CES 2026, Nvidia announced two major milestones that underscore its push to dominate both large-scale AI infrastructure and physical AI systems. Jensen Huang, Nvidia’s founder and CEO, confirmed that the company’s next-generation Rubin computing architecture is now in full production, with broader deployment expected in the second half of the year. Rubin succeeds Blackwell and is designed to meet surging compute, memory, and interconnect demands driven by advanced and agentic AI workloads. Major cloud providers and partners, including OpenAI, Anthropic, Amazon Web Services, HPE, and Lawrence Berkeley National Laboratory, are slated to deploy Rubin-based systems.
Alongside Rubin, Nvidia introduced Alpamayo, a new open family of AI models, simulation tools, and datasets aimed at autonomous vehicles and robotics. The release centers on Alpamayo 1, a 10-billion-parameter vision-language-action model built to reason through complex real-world driving scenarios. Led by Nvidia’s automotive team under Ali Kani, the platform combines open models, large-scale driving datasets, generative world models via Cosmos, and the new AlpaSim simulation framework.
Together, Rubin and Alpamayo position Nvidia at the core of both hyperscale AI compute and the emerging generation of reasoning-driven physical AI systems.
