Nvidia Unveils Blackwell Ultra and Vera Rubin Chips at GTC 2025

During the GTC 2025 keynote on March 18, Nvidia CEO Jensen Huang introduced significant advancements in the realm of artificial intelligence hardware with the announcement of two new chips: Blackwell Ultra and Vera Rubin. These innovations reflect Nvidia's commitment to addressing the soaring demand for AI infrastructure, particularly in data centers.

Blackwell Ultra: Features and Performance

Blackwell Ultra is a high-performance AI chip designed with 288GB of memory, which positions it as a formidable player in the AI accelerator market. Set to ship in the second half of 2025, this chip is expected to cater to the growing needs of companies leveraging AI for various applications. This announcement underscores Nvidia's strategy of rolling out advanced AI accelerators on an annual basis, anticipating and responding to the increasing data processing requirements of enterprises.

The Vera Rubin Platform

Alongside Blackwell Ultra, Jensen Huang previewed the Vera Rubin platform, which marks a significant leap in Nvidia's hardware capabilities. The platform features Nvidia’s first custom CPU, named Vera, in conjunction with the Rubin GPU. This pairing is engineered to target an impressive inference performance of up to 50 petaflops, making it a critical tool for heavy computational tasks. Scheduled for release in late 2026, Vera Rubin is indicative of Nvidia's foresight in AI technology, with a plan to follow it with the Rubin Ultra variant in 2027. This roadmap positions Nvidia to sustain its leadership in AI processing, amid the rapid evolution of data handling demands in modern computing environments.

Nvidia is addressing the increasing necessity for efficient and powerful AI accelerators, reflecting the company's robust roadmap tailored to future data center needs.

The announcements made at GTC 2025 are not only pivotal for Nvidia's product lineup but also highlight the broader trend in the tech industry toward relentless innovation in artificial intelligence capabilities. With these advancements, Nvidia continues to solidify its role at the forefront of AI hardware development.

For further details on these announcements, readers can visit CNBC and TechCrunch.