Nvidia revenue more than doubles as demand for AI chips remains strong

Nnvidia CEO Jensen Huang. Photo / Getty Images

Nvidia’s revenue more than doubled in the past quarter, continuing the US chipmaker’s run of blockbuster growth as demand soars for chips that can power the rollout of powerful artificial intelligence tools.

The Silicon Valley-based company sought to reassure investors that it would see “several billion dollars” in revenue this fiscal year from the next generation of its powerful AI chips, despite hitting production problems, even as its outlook for the current quarter fell shy of Wall Street’s most ambitious forecasts.

Revenue in the three months through July was US$30 billion, up 122% from a year ago. Analysts had expected US$28.7b. Nvidia is expecting US$32.5b in revenue for the current quarter, plus or minus 2%, only just ahead of analysts’ consensus expectations.

Some investors had been looking for an even higher revenue forecast in the run-up to Wednesday’s report. Shares fell about 6% in after-hours trading immediately after the release despite the strong earnings.

Nvidia also authorised another US$50b in share buybacks.

The hotly anticipated earnings report has been watched by investors for signs of how the artificial intelligence boom that has gripped the tech sector is faring.

While the year-on-year growth represented another record for Nvidia, it was far less than the 262% jump in revenue it had reported in the prior quarter. Earnings per share were 68 US cents, versus estimates for 65 cents. Its gross margin hit 75.1%, compared with analysts’ expectations for 75.5%.

Earlier this month, a delay to its next generation of chips, known as Blackwell, emerged as a potential hurdle to Nvidia’s breakneck growth. Chief executive Jensen Huang had previously said Blackwell would generate “a lot” of revenue for the company this year.

Nvidia’s finance chief, Colette Kress, appeared to reassure investors about that on Wednesday. She said Nvidia, which works with Taiwan Semiconductor Manufacturing Company to build the chips, had made changes to how Blackwell was produced to improve manufacturing yield.

She added: “Blackwell production ramp is scheduled to begin in the fourth quarter and continue into fiscal 2026. In the fourth quarter, we expect to ship several billion dollars in Blackwell revenue.”

Huang added that demand for its current-generation Hopper chips “remains strong”.

Nvidia has taken on an outsized importance in US stock markets, after a blistering rally pushed its shares up about 160% in the year to date, giving it a market capitalisation of US$3 trillion. Its growth has driven more than a quarter of the year-to-date gains on the S&P 500.

Investors have been braced for volatility around earnings, with options markets earlier this week pricing in a potential swing of 10% in either direction for the stock.

The latest quarterly results from Google, Microsoft, Meta and Amazon have shown the size of Big Tech’s spending spree to build the infrastructure to train and run AI models. They are also among Nvidia’s biggest customers, and the earnings report was expected to offer a temperature check on the broader mood around AI.

Daniel Newman, chief executive of Futurum Group, described it as a “solid quarter, but anything less than a guide to the top of estimates would likely be met with some consternation”.

“Blackwell will really hit in Q4, which may have surprised some but shouldn’t have. Hopper demand should allow the company to safely beat its [guidance], which was ahead of consensus but still conservative in my view,” he told the Financial Times.

Written by: Michael Acton in San Francisco

© Financial Times