Nvidia unveils next-generation Rubin AI chips as it surprisingly launches the new era of AI earlier than expected
Nvidia has unveiled its latest generation of artificial intelligence chips, just months after the previous model was unveiled, in a sign of the rapid progress and acceleration in AI technology.
The rapid turnaround underlines the fierce competition in the AI chip market and Nvidia’s drive to maintain its leadership not only as an AI chipmaker, but also as one of the world’s most valuable companies.
Ahead of the Computex 2024 technology conference in Taipei, Nvidia CEO Jensen Huang introduced the new AI chip architecture, called Rubin, as the successor to Blackwell.
Nvidia announces the Rubin AI chip just months after Blackwell
The introduction of Rubin AI chips comes just months after Nvidia unveiled its Blackwell model, which is still in production and expected to launch later this year.
Huang’s announcement surpasses the company’s new commitment to launch new AI chip technology once a year, which was already faster than the previous two-year cycle.
The Rubin platform includes new GPUs and an Arm-based CPU called Vera, along with NVLink 6, CX9 SuperNIC and the X1600 converged InfiniBand/Ethernet switch for advanced networking. Specific details surrounding the new Rubin chips have yet to be announced.
Nvidia’s decision to accelerate development underlines its strategy to stay ahead of rivals such as AMD and Intel. The company counts major technology companies such as Microsoft, Google and Amazon among its customers, even though the companies develop their own internal processors.
Since the announcement, Nvidia shares have held steady as the company continues to make big strides toward becoming the most valuable company in the world. With its current market capitalization of $2.695 trillion, Nvidia ranks third, behind Microsoft and Apple, both hovering around $3 trillion, and ahead of fourth-place Google, which is valued at $2.139 trillion.