LAS VEGAS, 5 January 2026 — NVIDIA Corporation chief executive officer Jensen Huang said the company’s next generation of chips is now in full production, marking a major step forward as demand for artificial intelligence computing continues to accelerate.
Speaking at the Consumer Electronics Show (CES) in Las Vegas on Monday, Huang said the new chips are capable of delivering five times the AI computing performance of Nvidia’s previous generation when powering chatbots and other AI applications.
Huang revealed that the chips, which are set to arrive later this year, are already being tested by AI firms in Nvidia’s laboratories, even as the company faces rising competition from both traditional rivals and its own customers.
The new Vera Rubin platform, which comprises six separate Nvidia chips, is expected to debut later this year. The flagship server configuration will contain 72 Nvidia graphics processing units and 36 new central processing units. Huang demonstrated how the systems can be linked into large-scale “pods” with more than 1,000 Rubin chips, significantly boosting efficiency.
According to Huang, the platform can improve the efficiency of generating AI “tokens”, the basic units used by AI systems, by as much as 10 times.
“This is how we were able to deliver such a gigantic step up in performance, even though we only have 1.6 times the number of transistors,” Huang said.
To achieve the performance gains, Huang said the Rubin chips rely on a proprietary data format, which Nvidia hopes will be adopted more broadly across the industry.
While Nvidia continues to dominate the market for training AI models, competition is intensifying in the deployment and serving of those models to large-scale users. Rivals include Advanced Micro Devices, as well as major customers such as Alphabet Inc., which are developing their own AI hardware.
A significant portion of Huang’s presentation focused on inference performance, where AI models deliver responses to users. Nvidia unveiled a new storage layer known as “context memory storage”, designed to help chatbots respond more quickly to long prompts and extended conversations.
Nvidia also introduced a new generation of networking switches featuring co-packaged optics, a technology critical for linking thousands of machines into a single computing system. The networking offering places Nvidia in direct competition with Broadcom Inc. and Cisco Systems.
