Nvidia Unveils Next Generation of AI Chips

Nvidia has announced a new chip architecture for training artificial intelligence models, named Rubin, as reported by CNBC.

The announcement comes two and a half months after the presentation of the previous generation AI processors, Blackwell. This chip is still in production and is expected to be available by the end of 2024.

Journalists have speculated that the accelerated pace of new technology development in the industry indicates heightened competition and Nvidia’s effort to maintain its dominant position.

Company CEO Jensen Huang stated that the firm plans to produce new AI chip models “on an annual basis.” Previously, the tech giant released one chip every two years.

“Today we are on the brink of a significant shift in computing. With our innovations in artificial intelligence and accelerated computing, we are pushing the boundaries of what is possible and launching the next wave of technological progress,” Huang added.

Rubin will consist of a new graphics processor and a central board called Vera. Detailed specifications have not yet been disclosed.

Following the announcement of the AI chip, Nvidia’s shares rose by 3.7% to $1139.

Previously, the company introduced the Latte3D model for generating 3D objects from prompts.

Back in January, Meta’s head Mark Zuckerberg promised to purchase Nvidia chips for the corporation, as the “future roadmap” in artificial intelligence requires “enormous computing infrastructure.”