Nvidia Reports Strong Q3 Earnings Amidst Blackwell Production Ramp-Up

Nvidia's Blackwell: All Systems Go?

Nvidia, the chip giant, recently announced its third-quarter results. Anticipation was high, particularly regarding the future of Artificial Intelligence (AI) and the role Nvidia plays within it. The focus was on the new AI chip, Blackwell, which shipped a few months late after production issues. Rumors of overheating problems when connecting multiple chips in Nvidia's custom-built server racks further fueled uncertainty.

However, Nvidia CEO Jensen Huang reassured analysts and investors during the earnings call. Blackwell production is running at full speed, and the company will ship more Blackwell chips this quarter than originally planned. Key customers like Microsoft and OpenAI have already received the new chips, and revenue from Blackwell could reach several billion dollars next quarter.

Industry analysts had previously reported that the problems with the new chips had been resolved. Dylan Patel, chief analyst at SemiAnalysis, told Business Insider that the overheating issues "have been known and largely solved for months." Patrick Moorhead, founder and chief analyst of Moor Insights & Strategy, also told WIRED that his contacts in manufacturing hadn't reported any major overheating problems. However, tensions between design, engineering, and production are not uncommon during the launch of new chip designs.

“It’s an age-old debate in the semiconductor industry: Is it a design problem? Is it a manufacturing problem?” Moorhead said. “People are looking for any little clue that would discredit Nvidia’s dramatic growth thesis.”

Nvidia generated $35.1 billion in revenue in the third quarter, surpassing estimates of $33.2 billion. Revenue increased 94 percent year over year. That’s a “small” increase compared to recent quarters, when revenue increased by as much as 265 percent year over year—but Nvidia is far from being in trouble. The results suggest the AI market continues to boom as companies invest billions of dollars in buying advanced chips and other equipment, albeit at a slightly slower pace.

Much of Nvidia’s growth this quarter was driven by data center revenue, which totaled $30.8 billion, up 112 percent year over year. The company’s gross margin was 74.5 percent, essentially flat year over year. However, analysts expect Nvidia’s margins could shrink as the company shifts to producing more Blackwell chips, which have higher manufacturing costs than their less advanced predecessors.

Nvidia's earnings reports are considered a key indicator for the entire AI industry. The chip architect’s advanced GPUs, which enable complex neural network processes, have made the current generative AI boom possible. As Silicon Valley giants raced to build new chatbots and image-generation tools in recent years, Nvidia’s revenue exploded, allowing the company to surpass Apple as the world’s most valuable publicly traded company. Since the launch of ChatGPT in November 2022, Nvidia’s stock price has nearly increased tenfold.

Almost every major tech company working on AI, even those building their own processors, relies heavily on Nvidia GPUs to train their AI models. Meta, for example, has stated that it’s building its latest AI technology on a cluster of more than 100,000 Nvidia H100s. Smaller AI startups, on the other hand, have been starved of AI compute power, as Nvidia has struggled to keep up with demand.

Blackwell, Nvidia's latest GPU, is made up of two pieces of silicon, each the size of the previous Hopper chip, combined into a single component. This design has resulted in a chip that is reportedly four times faster and has more than twice as many transistors as its predecessor.

But Blackwell’s launch hasn’t been entirely smooth. Originally slated for the second quarter, the new chip ran into production difficulties that reportedly delayed its release by a few months. Huang took responsibility for the issue, calling it a “design flaw” that “resulted in low yields.” Huang told Reuters in August that Nvidia’s longtime chipmaking partner, Taiwan Semiconductor Manufacturing Company Limited, had helped Nvidia fix the problem.

Moorhead told WIRED he remains bullish on Nvidia and is confident that the market for generative AI will continue to grow for at least the next 12 to 18 months, despite some recent reports suggesting that AI progress is stagnating.

“I think the only way shareholders would revolt is if they’re worried about the capex or profitability of the hyperscalers,” Moorhead said, referring to large tech companies like Amazon, Google, Microsoft, and Meta that are heavily invested in AI cloud services. “But I think they’ll just keep buying Nvidia until that day actually happens.” Enterprise AI is also a growth area for Nvidia, he added.

On today’s earnings call, Colette Kress, Nvidia’s chief financial officer, said that Nvidia’s enterprise AI tools are “firing on all cylinders,” including an operating platform that allows other businesses to develop their own copilots and AI agents. Customers include Salesforce, SAP, and ServiceNow, she said.

Huang echoed the same sentiment later in the call: “We’re slowly seeing the adoption of agentic AI in enterprises,” he said. “It’s really all the rage.”