Nvidia’s stock caught fire after a volatile close Wednesday, as investors refocused on a record quarter, a bigger-than-expected outlook and signs the AI buildout is still accelerating. The chipmaker said fiscal second-quarter revenue hit $96.2 billion, blowing past a year-earlier level that was 106% lower and rising 18% from the prior quarter, while third-quarter sales are now projected at $108 billion, plus or minus 2%. Shares finished the regular session at $209.66, down 1.59%, before turning higher in after-hours trading and then jumping again in premarket action Thursday.
The market reaction tells the story: this is no longer just a results beat, it is a test of whether the AI spending cycle still has room to run. Nvidia’s numbers suggest the answer remains yes, even with investors already braced for another blockbuster. The company’s Data Center business, the engine behind the AI trade, posted $89.0 billion in revenue, up 117% from a year earlier. That kind of growth, at this scale, is rare even in the most euphoric phases of a market cycle.
For a company already worth more than most rivals can imagine, Nvidia just posted one of the biggest quarterly revenue figures in corporate history. Fiscal Q2 revenue of $96.2 billion came in alongside non-GAAP earnings of $2.22 a share, and GAAP earnings of $2.46 a share. Gross margin was 75.0% on both measures. The results show that the company is not only shipping more chips, but doing so at pricing and profitability levels that keep investors willing to pay up for the stock.
Guidance mattered just as much. Nvidia’s forecast for fiscal Q3 revenue at $108 billion, plus or minus 2%, points to another step up from already enormous levels. That estimate also sits above the Wall Street figure cited by Reuters in the source material, adding to the perception that demand is still outrunning even elevated expectations. The company also said the outlook assumes no Data Center chip sales to China, leaving a potential upside if shipments into that market improve.
Founder and CEO Jensen Huang framed the results as part of a much larger economic transition around AI. He said, “AI has reached its inflection point. It’s doing useful work. Its tokens are productive and profitable. Now, compute is revenue. And demand is accelerating.” The language matters because it links Nvidia’s business directly to the monetization of AI, not just the hype around it. Huang is arguing that the spending wave is now justified by commercial output, and the latest quarter gives that pitch more credibility.
That message is especially powerful for a market that has repeatedly asked whether the AI trade is mature or merely pausing. Nvidia’s report says the cycle is still broadening. The company pointed to demand across frontier labs, startups, open-model developers and physical AI, suggesting the customer base is widening rather than narrowing. In practical terms, that means the story is no longer just about a few large cloud buyers; it is becoming a much larger infrastructure race.
Nvidia also highlighted another large expansion with Amazon Web Services, underscoring how central the company has become to hyperscale cloud buildouts. The source material says Nvidia and AWS plan to deploy another 2 million GPUs across Amazon’s global infrastructure during 2027 and 2028. That is a staggering figure even by AI standards, and it shows the scale of the commitments still flowing into the sector.
The AWS announcement matters because it helps explain why the stock bounced back so quickly after a weaker regular-session finish. Investors are not just reading a single quarter. They are looking at the next several years of demand visibility. When a company with Nvidia’s market weight says it is preparing for another 2 million GPUs to be deployed through a major cloud partner, the market tends to treat that as a signal that the capex cycle remains alive and well.
Nvidia said its Vera Rubin platform is ramping into full production, with racks running at CoreWeave, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure and Nebius. That detail matters because it shows the next product cycle is not just on a slide deck. It is already entering deployment across a range of major cloud and infrastructure players. For a company that has spent the past several quarters turning AI demand into revenue, the Rubin ramp suggests the pipeline is being replenished rather than emptied.
But the scale of the opportunity comes with a bigger financial burden. Nvidia disclosed $366 billion in multiyear AI infrastructure commitments, including $279 billion tied to supply and capacity, along with up to $108.5 billion in phased guarantees. Those numbers point to a business model that is becoming more capital intensive as the AI ecosystem expands. Nvidia is helping finance and secure the infrastructure that powers the boom, which deepens its role in the market while also increasing its exposure to execution risk.
That combination is part of what keeps the stock so tightly linked to sentiment around AI spending. If hyperscalers, cloud platforms and AI developers keep building, Nvidia benefits on several fronts at once: chip sales, system demand and the optics of being the indispensable supplier. If spending slows, those same commitments could look heavier. For now, the company’s report suggests the first scenario still has the upper hand.
The stock reaction was noisy, but the direction was clear. Nvidia closed regular trading down 1.59% at $209.66, then reversed after the earnings release. Market data in the source material shows the shares later rose more than 4% after hours, and premarket trading on Thursday put the stock up about 7.38% at $225.13. Different time windows explain the conflicting headlines around the move, but the takeaway is the same: investors liked what they heard once they absorbed the size of the beat and the strength of the guide.
The company also told shareholders it returned about $26.0 billion and still had about $99.0 billion remaining under its buyback authorization. That gives management another lever to support the stock, especially at a moment when the market is already rewarding scale, execution and cash generation. Nvidia is also set to pay its next quarterly dividend of $0.25 a share on October 1, 2026, to shareholders of record on September 10, 2026.
What comes next is whether Nvidia can keep converting the AI frenzy into even bigger numbers without losing momentum. The company has already delivered record revenue, record non-GAAP EPS and a guide that keeps rising. It also said it sees roughly 70% long-term revenue growth for fiscal 2028. For a firm this large, that is a message the market cannot ignore: the AI spending boom is not just intact, it is still being upgraded.
