Jensen Huang, Nvidia Corp. co-founder and CEO, during the Nvidia GPU Technology Conference (GTC) on Tuesday, March 19, 2024 in San Jose, California, USA.
David Paul Morris | Bloomberg | Good pictures
Nvidia unveiled its next-generation artificial intelligence chips on Sunday, following the previous model announced in March.
Nvidia CEO Jensen Huang It announced a new AI chip architecture called “Rubin” ahead of the COMPUTEX technology conference in Taipei.
It comes months after the March announcement of Rubin’s upcoming “Blackwell” model, which is still in production and expected to ship to customers in 2024.
Huang’s announcement about Rubin accelerates the company’s already accelerating pace of AI chip progress.
As Huang said on Sunday, Nvidia has committed to releasing new AI chip models in a “one-year rhythm.” The company previously operated on a slow two-year refresh schedule for chips.
The switch from Blackwell to Rubin is less than three months old, underscoring the frenzy of competition in the AI chip market and Nvidia’s speed to defend its dominant position.
AMD and Intel are the two main competitors, although their gross margins lagged behind Nvidia’s in the most recent fiscal quarter. Companies like Microsoft, Google and Amazon are also vying for Nvidia’s top spot. A flurry of startups are also working to enter the space.
“Today, we are on the cusp of a major transformation in computing,” Huang said Sunday. “Through our innovations in AI and accelerated computing, we’re pushing the boundaries of what’s possible and driving the next wave of technological advancement.”
The Rubin chip platform will include new GPUs, a critical graphics processing technology that helps train and launch AI systems. While Sunday’s announcement did not provide many details, it will come with other new features such as a central app called “Vera”.
Shares of Nvidia were relatively flat at Friday’s market close, with shares trading at $1,096.
