Artificial-intelligence semiconductor powerhouse Nvidia on Monday unveiled a flagship AI chip, the Blackwell B200, saying it’s up to 30 times speedier than its previous chip.
Chief Executive Jensen Huang, kicking off the corporate’s annual developer conference, also launched a recent set of software tools designed to help developers sell their artificial-intelligence models more easily to any company that uses Nvidia.
Nvidia’s chip and software announcements at GTC 2024 will help determine whether the corporate can maintain its leadership position as the dominant seller of AI equipment. Nvidia had a roughly 80% share of the information center AI chip market last yr.
“I hope you realize this is just not a concert,” Huang, wearing his trademark black leather jacket, said after taking the stage, in a nod to the rising profile of his company.
The B200 takes two chips the dimensions of Nvidia’s previous offering and binds them together right into a single chip.
The brand new chip has 208 billion transistors, greater than double the 80 billion on the corporate’s previous chip. All of those transistors can access the memory attached to the chip at nearly the identical time, improving productivity.
Tom Plumb, CEO and portfolio manager at Plumb Funds, which has Nvidia as one among its largest holdings, said the Blackwell chip was not a surprise.
“However it reinforces that this company remains to be on the leading edge and the leader in all graphics processing. That doesn’t mean the market is just not going to be sufficiently big for AMD and others to are available in. However it shows that their lead is pretty insurmountable,” said Plumb.
Insider Intelligence analyst Jacob Bourne said Nvidia could solidify its AI dominance. “Nonetheless, rivals like AMD, Intel, startups, and even Big Tech’s own chip aspirations threaten to chip away at Nvidia’s market share, particularly amongst cost-conscious enterprise customers,” he said.
Nvidia said major customers, including Amazon, Alphabet’s Google, Meta Platforms, Microsoft, OpenAI, Oracle and Tesla, are expected to use the brand new chip.
Nvidia is also shifting from selling single chips to selling total systems. Its latest iteration houses 72 of its AI chips, 36 central processors and incorporates 600,000 parts and weighs 3,000 kilos.
Though Nvidia is widely regarded as a chip designer, the corporate has built a major battery of software products as well.
The brand new software tools, called microservices, improve system efficiency across a wide range of uses, making it easier for a business to incorporate an AI model into its work, just as a very good computer operating system may help apps work well.
Nvidia’s shares have surged 240% over the past 12 months, making Nvidia the US stock market’s third-most useful company, behind only Microsoft and Apple.
Nvidia stock dipped 1% in prolonged trade on Monday, while Super Micro Computer, which makes AI-optimized servers with Nvidia’s chips, fell 4%. Advanced Micro Devices stock dipped nearly 3% during Huang’s keynote address.
Its stellar 12-month rally leaves Nvidia’s stock liable to plummeting back to earth if the Santa Clara, Calif., company fails to expand its AI business as much as investors expect.
Nvidia’s market share is predicted to drop several percentage points in 2024 as recent products from rivals such as Intel and Advanced Micro Devices hit the market.
Huang spoke at a Silicon Valley hockey arena to accommodate its largest crowd for its annual conference.
Huang also announced partnerships with design software corporations Ansys, Cadence and Synopsys. Shares of the three corporations jumped around 3% in prolonged trade following Huang’s comments.