Nvidia founder and CEO Jensen Huang expressed confidence during the company’s third-quarter earnings announcement, and the numbers back that up.
For the third quarter, Nvidia reported revenues of $57 billion, marking a 62% increase compared to the same period last year. The company’s net income, calculated on a GAAP basis, was $32 billion, reflecting a 65% year-over-year rise. Both the revenue and profit figures surpassed Wall Street expectations.
The surge in revenue primarily stems from Nvidia’s booming data center business. This segment generated a record $51.2 billion, a 25% increase from the previous quarter and a 66% jump from a year ago. The remaining $5.8 billion in revenue came from Nvidia’s gaming sector, which brought in $4.2 billion, alongside contributions from professional visualization and automotive sectors.
Nvidia’s CFO, Colette Kress, highlighted in a statement to shareholders how the data center business has been driven by advancements in computing, powerful AI models, and innovative applications. During the Q3 earnings call, she mentioned that the company announced AI factory and infrastructure initiatives totaling an impressive 5 million GPUs.
“This demand stretches across all markets, including cloud service providers (CSPs), governments, modern enterprise builders, and supercomputing centers, and encompasses several large-scale projects,” Kress stated.
The recently launched Blackwell Ultra GPU, which comes in multiple configurations, has been particularly popular and is now leading the market for Nvidia. Previous iterations of the Blackwell architecture continue to enjoy strong demand, as confirmed by the company.
Huang stated, “Sales of our Blackwell GPU chips are off the charts.”
“Blackwell sales are extraordinary, and our cloud GPUs are sold out,” Huang noted in the Q3 earnings statement. “Compute demand is rapidly accelerating across both training and inference, each growing exponentially. We have entered a virtuous cycle of AI. The AI ecosystem is expanding quickly—with more foundation model developers, more AI startups, and involvement across more industries and countries. AI is spreading widely and impacting everything simultaneously.”
Kress acknowledged, however, that shipments of H20, a data center GPU aimed at generative AI and high-performance computing, totaled 50 million—a figure that fell short due to challenges in selling to China.
“Large purchase orders didn’t materialize this quarter because of geopolitical factors and increasing competition in China,” Kress mentioned during the call. “While we are disappointed by the current situation that limits our ability to ship competitive data center products to China, we remain committed to ongoing discussions with the U.S. and Chinese governments to support America’s global competitiveness.”
Looking ahead, Nvidia anticipates continued growth, projecting fourth-quarter revenue of $65 billion, which helped boost its share price by over 4% in after-hours trading.
In Huang’s perspective, the focus should be on growth rather than concerns about an AI bubble.
“There’s a lot of chatter about an AI bubble,” he remarked during the earnings call. “From our point of view, we see something quite different.”
