Nvidia Announces Next-Generation AI Chip Platform for 2026 Rollout
Nvidia unveils Rubin, its next-generation AI processor platform, which will be available in 2026. The Rubin chip family will comprise new GPUs, CPUs, and networking components. The new CPU will be named Versa, while the graphics processors will use next-generation high-bandwidth memory.
Nvidia, the world's largest manufacturer of artificial intelligence (AI) processors, has announced its ambitions for the next-generation AI chip platform. During a presentation at the Computex trade event in Taipei, Nvidia CEO Jensen Huang announced that the future platform, dubbed Rubin, will be available in 2026.
The Rubin family of chips will comprise a variety of components, including new graphics processing units (GPUs), central processors (CPUs), and networking chips. While Huang did not go into much detail on the platform, he did reveal that the new CPU will be dubbed Versa. Furthermore, the graphics processors will use next-generation high-bandwidth memory supplied by industry leaders such as SK Hynix, Micron, and Samsung.
One remarkable feature of Nvidia's statement is the company's intention to introduce a new family of AI processors every year, a huge increase from its prior release cadence of once every two years. This plan demonstrates Nvidia's dedication to remaining at the forefront of AI technology while satisfying the market's expanding demands.
Nvidia's strong position in the AI chip business, which controls over 80% of the industry, positions it as both the main enabler and beneficiary of AI's rapid development. The company's chips power a wide range of applications, from self-driving cars to data centres, and its continuing innovation is critical to propelling AI technology forward.
Nvidia announces next-generation AI chip platform, Rubin, to be rolled out in 2026
The Rubin family of chips will include new GPUs, CPUs, and networking chips
The new CPU will be called Versa, and the graphics chips will feature next-generation high-bandwidth memory
Source: REUTERS