Nvidia Corporation co-founder and CEO Jensen Huang at the Nvidia GPU Technology Conference (GTC) in San Jose, California, USA, on Tuesday, March 19, 2024.
David Paul Morris | David Paul Morris Bloomberg | Getty Images
Nvidia On Sunday, the company unveiled its next-generation artificial intelligence chip to succeed the previous model, which was announced a few months ago in March.
CEO of NVIDIA Jensen Huang A new AI chip architecture called “Rubin” was announced before the COMPUTEX technology conference in Taipei.
Rubin’s announcement comes just months after the March announcement of the upcoming “Blackwell” model, which is still in production and expected to ship to customers later in 2024.
Jen-Hsun Huang’s announcement of the launch of Rubin appears to have accelerated the company’s already accelerated pace of artificial intelligence chip research and development.
As Huang said on Sunday, Nvidia has promised to release new artificial intelligence chip models at a “one-year pace.” The company’s previous chip refresh schedule was slower at two years.
The transition from Blackwell to Rubin, which took less than three months, highlights the fierce competition in the AI chip market and Nvidia’s sprint to maintain its dominance.
AMD and Intel Nvidia is the two main rivals struggling to catch up, although their gross margins trailed Nvidia’s in the most recent quarter. company likes Microsoft, Google and Amazon Although they are also one of Nvidia’s largest sponsors, they are also competing for Nvidia’s top spot. Many new startups are also trying to enter this field.
“Today, we are on the cusp of a major shift in computing,” Huang said Sunday. “With our innovations in artificial intelligence and accelerated computing, we are pushing the boundaries of what is possible and driving the next wave of technological advancements.”
The Rubin chip platform will be equipped with new GPUs, a key graphics processing technology that helps train and launch artificial intelligence systems. It will come with other new features, such as a central processing unit called “Vera,” although Sunday’s announcement didn’t offer many details.
Nvidia shares were relatively flat as of Friday’s close, trading at $1,096.