Nvidia Corp. (NVDA) just delivered the kind of earnings report that makes competitors rethink their life choices. The chip giant posted third-quarter results Wednesday evening that didn't just beat Wall Street expectations—they steamrolled them—and the guidance that followed made it clear that Nvidia's lead in AI hardware isn't shrinking anytime soon.

The numbers tell the story: $57.01 billion in revenue versus the $54.88 billion analysts were expecting. Earnings per share came in at $1.30, topping the $1.25 consensus and representing a 60.5% jump from last year. Gross margin? A staggering 73.6%. Operating income reached $37.75 billion, translating to an operating margin of 66.2%—numbers that put Nvidia among the most profitable companies in the entire S&P 500.

But here's the thing: the beat itself wasn't even the main event. What really caught Wall Street's attention was what Nvidia said about the future.

Data Centers Are Printing Money

The real engine behind Nvidia's growth continues to be its Data Center segment, which generated $51.2 billion in revenue—up 56% from a year ago and more than $1.5 billion above what analysts expected. The AI infrastructure buildout is very much alive and well.

Goldman Sachs analyst James Schneider noted in a Thursday research note that Nvidia's management addressed "several key investor questions" that had been weighing on the stock recently. More importantly, they provided evidence suggesting upside to the company's previous outlook of $500 billion in Data Center revenue for 2025-2026.

"Nvidia sees potential upside to its prior Data Center revenue outlook of $500 bn in 2025/26," Schneider wrote. Translation: that massive number might actually be conservative.

And it's not just about selling chips anymore. Nvidia's networking business—which includes NVLink, SpectrumX, and Infiniband—pulled in $8.2 billion in revenue, up 162% year-over-year. The customers fueling that growth read like a who's who of tech: Meta Platforms Inc. (META), Microsoft Corp. (MSFT), Oracle Corp. (ORCL), and Elon Musk's xAI are all building out massive AI infrastructure, and they're all buying Nvidia's gear to connect it all together.

Guidance That Keeps Getting Better

Nvidia guided fourth-quarter revenue to $65 billion—comfortably ahead of Goldman's $63.2 billion estimate and well above the Street's $62.4 billion consensus. The company also guided gross margins to 75% and implied earnings per share of $1.50, both topping expectations.

"Nvidia believes it can hold gross margins in the mid-70% range as higher pricing and other cost reductions offset these increased input costs," Schneider explained. That's crucial because maintaining those margins while scaling production isn't easy, but Nvidia seems to have figured it out.

Schneider's take? "We see significant upside to Street estimates, and we view valuation as relatively appealing at current levels."

Based on that conviction, Goldman Sachs raised its 12-month price target from $240 to $250, reflecting increased confidence in Nvidia's earnings power and ability to maintain premium margins. But that's just the base case.

Three Ways This Could Play Out

Goldman laid out three scenarios, and the range is wide. In the base case, the firm expects Nvidia to trade at a 30x forward earnings multiple on revised next-twelve-month EPS of $8.25, yielding that $250 price target—a 27.1% upside from current levels.

The bull case is where things get interesting. If AI infrastructure spending accelerates faster than expected and Nvidia captures even more market share across data center workloads, Goldman sees earnings hitting $9.50 per share. Apply a 35x multiple to that, and you get a $333 price target—implying nearly 70% upside.

The bear case assumes the AI spending boom cools off, competition intensifies, and margins compress. Under that scenario, with $5.80 in projected EPS and a 25x multiple, the price target drops to $145, representing a potential 26.3% downside.

Perhaps most telling, Goldman introduced long-term EPS projections for the first time: $15.60 in 2028, $18.65 in 2029, and $22.10 in 2030. Those aren't numbers you throw out casually—they signal Goldman believes Nvidia's dominance extends well beyond the current product cycle.

Schneider summed it up nicely: "We believe Nvidia has a sustainable model advantage over peers in AI training applications." Right now, the AI race might have other contestants, but there's only one horse worth betting on.