Nvidia posted record earnings, beating analyst expectations while brushing aside circular financing claims and shrugging off emerging chip manufacturer concerns.
The company posted highest ever revenue of $96.2 billion during its fiscal second quarter 2027, which ended July 26, up 106% year over year, according to the its earnings reports. Of that revenue, $89 billion, or 92%, is attributed to data center build-out.

“We expect to grow revenue by approximately 70% in fiscal 2028,” Chief Executive Jensen Huang said during the Q2 earnings call on Aug. 26.
Another part of that revenue is compute, Huang said.
Earlier this year, Nvidia launched a new GPU, Vera Rubin, which delivers 30 times higher throughput per megawatt and 35 times lower token cost compared to its predecessor, the Blackwell Ultra generation GPUs that were launched in March 2025, Huang said.
Circular economy
Nvidia sells GPUs, but also offers AI financing through its recent partnerships with Apollo, Blackrock, Blackstone, Brookfield, KKR and Goldman Sachs, according to an Aug. 10 release.
In an Aug. 26 credit memo released before the earnings call, Morgan Stanley pointed to Nvidia’s AI financing as a circular economy, calling the latest financing move a “balance-sheet-as-a-service.”
Nvidia Chief Financial Officer Colette Kress pushed back on those claims during the earnings call.
“Independent capital still underwrites every deal on its own merits,” she said. “We’re not making loans.
“In this model, we get paid twice,” Kress said — once on the hardware sale and again when compute is used.
Chip manufacturing competitors
Other chip manufacturers are also innovating. Hypserscalers launching chips include:
- OpenAI, which unveiled Jalapeno, its first in-house accelerator build with Broadcom, this week;
- Google, which rolled out its own customer silicon, TPU v7, earlier this year.
Even with emerging chips hitting the market, Huang wasn’t rattled.
He said Nvidia’s chips still offer two to eight times better performance per watt than application-specific chips.
“There are a lot of chips that get built” that never get deployed at scale, he said.forowever, the stakes are significant because Nvidia’s customer base is roughly half hyperscalers — accounting for 70% of its revenue, according to the company’s 10-Q filed with the SEC on Aug. 26.
High demand, but somehow high inventory
To reach its 70% revenue growth goal for fiscal 2028, Nvidia insists supply, not demand, is the constraint.
“Even though our demand is much greater than 70%, our supply allows us to confidently deliver 70%,” Huang said. “We’re going to continue to work with our supply chain to increase on that.”
Yet inventory keeps climbing. Nvidia’s stockpile hit $32 billion in Q2, more than double the $15 billion it held a year earlier — its sixth straight quarter of sequential growth, even as executives insist demand is outrunning what they can ship.
Shares of Nvidia (NYSE: NVD) were up 8.4% from market open to $227 as of market close today. The company has a market capitalization of $5.5 trillion.
Register here for the FinAi Lending Summit, set for Oct. 7-8 in Las Vegas. This inaugural event will include speakers from Fifth Third and Capital One as well as a fireside chat with Piermont Bank founder and Chief Executive Wendy Cai-Lee.





