Nvidia went into the quarter carrying the weight of an entire market narrative, and it walked out with numbers that could have analysts redrawing all of their models
Photo by Jonathan Raa/NurPhoto via Getty Images
Nvidia $NVDA walked into earnings carrying the weight of a monthlong AI panic and walked out with a $57 billion quarter that made the whole selloff look like stage fright. The data-center business alone grew into a $51 billion engine, the kind of gravitational force that reorganizes the rest of the sector around it.
The market didn’t take long to recalibrate: Stock was up as much as 6.4% in after-hours trading. The print read like a reminder that the biggest constraint in AI remains supply, not sentiment, and traders responded the way they usually do when an idea survives contact with reality. Wedbush’s Dan Ives called it a “champagne” moment for a reason; screens tilted green once the numbers hit.
Wall Street had braced for something good, but not necessarily something quite this loud. Analysts were looking for about $55.4 billion in revenue and $1.26 a share in earnings. Instead, Nvidia cleared both bars without breaking cadence. Revenue hit $57 billion (up 62% from a year ago and 22% from the prior quarter), earnings landed at $1.30 a share, and the growth rate pushed the entire conversation back toward scale rather than saturation. Even the margins — 73.4% on a GAAP basis — held in place, a small-but-critical signal for a company ballooning this quickly.
But perhaps the biggest sign for Wall Street came from the outlook. Nvidia told investors to expect $65 billion in revenue next quarter, a number that jumped well past the Street’s $61.7 billion range and landed like a statement. If demand was slipping, Nvidia wouldn’t be leaning into the quarter with targets that size or gross-margin expectations edging toward 75% on a non-GAAP basis.
The beat and the guidance effectively imposed a new baseline for what “normal” demand looks like in this phase of the cycle. And the rest of the print drew the same shape.
Demand for Nvidia’s Blackwell-series chips has been running ahead of available supply for months, and early readouts from the call suggested that the pace hasn’t slowed. Cloud providers have been booking capacity across compute and networking to get ahead of Rubin’s ramp next year, turning the upgrade path itself into a catalyst. Nvidia CEO Jensen Huang said in the earnings release that “Blackwell sales are off the charts, and cloud GPUs are sold out.”
Compute and networking inside the data-center segment turned into a $51 billion pillar, with networking alone climbing more than 160% year-on-year as hyperscalers kept stitching together larger and larger AI clusters. Inventory rose to $19.8 billion because Nvidia is booking supply as quickly as manufacturers can produce it. Even the cloud-service agreements — where Nvidia effectively rents its hardware back through partners — doubled to $26 billion, a sign that customers are willing to lock in capacity before the next wave of model training crowds the queue. For all the hand-wringing about an AI cool-off, the spending news out of hyperscalers pointed in the opposite direction.
Heading into Nvidia’s earnings, some investors had been jittery. Major funds such as SoftBank and Peter Thiel trimmed their stakes earlier this month, and some analysts described Nvidia’s print as one of the key tests of whether the AI boom still has air.
Wedbush called it an “eye-popping” guide that rewrites investor sentiment; the firm’s early read on the call flagged strong tones around that Blackwell demand and Rubin’s ramp, perhaps enough to quiet the AI bears who spent November arguing that the upgrade cycle was already wobbling.
China still hovers in the background as the one conspicuous absence. Nvidia is still locked out of shipping its highest-end parts into the mainland, and the quarter reflects none of that demand. Any thaw in the trade negotiations expected next year could reopen a meaningful revenue stream, which means Nvidia is guiding to $65 billion without one of its historically important markets in the mix.
The physical limits on growth drew just as much attention. Power shortages, land constraints, and grid bottlenecks have been the running worry across hyperscalers, and analysts have pointed to those choke points as the next hard governor on AI spending. Nvidia didn’t wave the issue away, but the scale of the quarter showed what happens when customers lean into the buildout anyway.
Inventory climbed because the company is locking in supply months ahead, and long-term cloud commitments doubled because clients want guaranteed access before the next wave of model training crowds the system. The entire setup carried the feel of a market pushing against infrastructure rather than demand.
Join 500,000+ readers who start their day with Quartz.
By subscribing, you agree to our Terms of Service and Privacy Policy.
