Nvidia showcased what he described as physical artificial intelligence (AI) at the Consumer Electronics Show in Las Vegas on Monday (Jan. 5).
Nvidia CEO Jensen Huang said the company’s next generation of chips can deliver five times the amount of AI computing compared to its previous chips.
“Today, I can tell you that Vera Rubin is in full production,” Huang said.
Physical AI refers to systems that operate in real environments, such as robots and self-driving cars.
The company says the robots are designed to perceive, decide, and act in the real world.
Powered by Nvidia chips and software, the robots bring advanced intelligence into action.
According to Nvidia, physical AI systems need to understand space, predict outcomes, and respond safely in real time.
The company says it requires far more training than AI software that’s currently on the market.
Set to launch later this year, the flagship Vera Rubin platform packs six separate Nvidia chips with 72 graphic units and 36 new central processors.
These can be linked into pods, with over 1,000 Rubin chips working together.
To achieve these performance gains, Huang said the hardware uses proprietary data formats that he hopes the industry will embrace.
“This is completely revolutionary. This is how we were able to deliver such a gigantic step up in performance, even though we only have 1.6 times the number of transistors,” he said.
The AI chip leader also introduced new features aimed at helping chatbots handle complex conversations with millions of users simultaneously.
Huang also showcased new self-driving car software, called Alpamayo, that he said will be open-sourced.
“Not only do we open-source the models, we also open-source the data that we use to train those models. Because in that way, only in that way can you truly trust how the models came to be,” he said.
Nvidia’s Competitor Advanced Micro Devices Showcases Advanced AI Processors
The same day saw rival Advanced Micro Devices (AMD) show off its latest AI chips in Las Vegas too.
“MI455 is the most advanced chip we’ve ever built. It’s pretty darn big,” ADM CEO Lisa Su said. “It has 320 billion transistors, 70 per cent more than MI355.”
These chips power data centre racks that AMD sells to firms like OpenAI.
The president of ChatGP team maker, Greg Brockman, joined Su on stage.
He said chip advances were vital to OpenAI’s vast needs for computing power.
Su also had big claims for another platform.
“This is the smallest AI development system in the world, capable of running models with up to 200 billion parameters locally, not connected to anything,” she said.
AMD remains Nvidia’s top rival in AI chips, but trails in market share.
A recent deal with OpenAI has been seen as a show of confidence in the company.
