At the recent Computex conference in Taipei, Nvidia CEO Jensen Huang introduced the Rubin platform, marking a significant step in the company's rapid innovation cycle. Named after the esteemed astronomer Vera Rubin, this new AI platform follows closely on the heels of the Blackwell chip unveiled just a few months ago.
Rubin is designed to leverage HBM4, the latest evolution in high-bandwidth memory, and will feature a new CPU named Versa. The platform underscores Nvidia's commitment to accelerating its release schedule, with plans to launch a new family of AI chips annually—a notable increase from its previous two-year cycle. The Rubin family will include advanced graphics (GPU), central processors (CPU), and networking chips, with the Rubin Ultra set to follow in 2026.
Nvidia's aggressive expansion reflects its strategic push to diversify its customer base amid surging AI demand. The company's processors have become pivotal in the AI arms race among tech giants, contributing to billions in revenue. Huang emphasized that Nvidia is extending its reach to support a wide array of industries, including shipbuilding, pharmaceuticals, and government agencies, as they increasingly adopt AI technologies.
"We are seeing computation inflation," Huang remarked, highlighting the growing computational demands across sectors.
Nvidia's dominance in the AI chip market, where it holds an 80% share, positions it as a key player in the future of AI development. The introduction of Rubin not only demonstrates Nvidia's technological prowess but also its strategic vision to remain at the forefront of AI innovation.