Nvidia on Monday unveiled new chips for desktop and laptop PCs that use the same Blackwell architecture that powers the company’s fastest AI processors for servers and data centers.

The chips, called GeForce RTX 50 series, will be available pre-installed in computers for around $550 to $2,000, according to the company. Laptops with these chips will begin shipping in March.

Nvidia unveiled the processors at CES in Las Vegas, where CEO Jensen Huang gave a keynote speech on Monday.

“Can you imagine that you have this incredible graphics card, Blackwell? I’ll shrink it down and put it there,” Huang said, holding up a laptop.

Nvidia, whose market capitalization has grown to over $3.5 trillion by selling AI chips to giant cloud providers and other technology companies, was known until recent years for selling graphics processing units (GPUs) for video games. Nvidia’s first chip in late 1999 was designed to quickly draw triangles and polygons for 3D games.

“Of course, we were a gaming company back then and these GPUs were designed to accelerate games,” said Justin Walker, senior director of product at Nvidia, in a press call.

With the explosion of AI development and ever-increasing demand for more computing power, Wall Street is less enthusiastic about Nvidia’s gaming business these days. In the quarter that ended in October, Nvidia’s gaming sales accounted for less than 10% of total revenue, compared to 88% for data center chips.

Nvidia has the vast majority of the data center AI GPU market, outperforming the competition Advanced micro devices And Intel.

But CES, formerly known as the Consumer Electronics Show, is all about consumer products, and the new chips announced Monday are aimed primarily at gaming.

According to Nvidia, the RTX 50 series chips will support a feature called DLSS 4, which uses AI to boost frame rates in games. They can also display characters’ faces in more detail and generally provide users with better graphics and higher resolution.

Nvidia’s gaming business is growing, with revenue rising 15% year-over-year in its most recent quarter. But data center revenue has at least doubled for six consecutive quarters and most recently topped $30 billion.

Nvidia says technical improvements made for its massive AI business will also trickle down to its gaming graphics cards.

“While we are now both an AI company and a gaming company, our gaming side still benefits tremendously from the fact that we are an AI company,” Walker said.

The Blackwell GPU architecture and core design, which uses the 50-series chips, was first used in the company’s AI accelerators, which were announced in March and shipped later last year. Nvidia said they were designed and optimized to power neural networks used by OpenAI’s ChatGPT and Googles Gemini.

The new chips for PCs and laptops will be available in various configurations. The company says its most expensive and powerful chip, the RTX 5090, sells individually for $1,999 and is twice as fast as its predecessor, the RTX 4090. Nvidia says it has 92 billion transistors, versus 208 billion transistors at the company’s B200 GPU for servers.

Nvidia says the chips will be optimized for running AI models and computer graphics, not just running the latest games. The chips will be powerful enough for some game makers to integrate generative AI into their characters in games like “PUBG: Battlegrounds.”

The new processors will also be powerful enough to power large language models and image generation models from companies such as: B. to carry out MetaMistral and stability AI, Nvidia said.

REGARD: Bernstein’s Rasgon says investors need Nvidia exposure