Nvidia changes revenue reporting to focus on data centers and edge computing

Nvidia is ditching its old revenue reporting structure. Going forward, the company will break out its financial results by “Data Center” and “Edge Computing” instead of the previous “Compute and Networking” categories.

It’s the kind of accounting change that sounds boring until you realize it’s basically a corporate confession. Nvidia is telling Wall Street, in the clearest possible terms, that it’s no longer a GPU company that happens to sell to data centers. It’s a data center company that happens to make GPUs.

The numbers that forced the rebrand

Look at where the money is actually coming from. Nvidia’s data center revenue hit $51.2B in Q3 of fiscal year 2026. That’s roughly 90% of the company’s total revenue of $57B for the quarter.

When nine out of every ten dollars you earn comes from one business line, you don’t bury it inside a broader category called “Compute and Networking.” You put it front and center.

The growth trajectory here is staggering. Data center revenue essentially doubled, climbing from $30.77B in the prior-year period to $62.31B more recently. Fiscal 2025 full-year revenue landed at $130.5B, up 114% year-over-year, with AI data center demand doing most of the heavy lifting.

In English: Nvidia’s data center business is growing so fast that lumping it in with other product lines was starting to obscure more than it revealed. The old reporting structure was like tracking Amazon’s revenue by combining AWS with its book sales. Technically accurate, practically useless.

Why edge computing gets its own line

The other half of the new reporting structure, edge computing, is the more forward-looking play. Edge computing refers to processing data closer to where it’s generated, think autonomous vehicles, industrial robots, smart retail systems, rather than shipping everything back to a centralized cloud.

Nvidia has been building out its edge portfolio for years. Its Jetson platform targets robotics and embedded AI. Its DRIVE platform powers autonomous vehicle systems. Its Metropolis framework handles video analytics at the edge. None of these are small bets.

By giving edge computing its own reporting segment, Nvidia is signaling that this isn’t a side project. It’s the company’s next growth vector, the business line it expects investors to track quarter over quarter as it scales.

Here’s the thing. Data centers are where the money is today. Edge computing is where Nvidia wants the money to come from tomorrow. The new reporting structure makes both of those stories visible in a way the old one simply didn’t.

What this tells us about the AI infrastructure race

This restructuring arrives at a moment when Nvidia’s position in AI infrastructure is both dominant and contested. The company’s guidance suggested Q3 FY2026 revenue of approximately $54B, which actually came in below some analysts’ more optimistic projections. Even when Nvidia beats expectations, the goalposts keep moving.

The competitive pressure is real. AMD is pushing its MI300 series as a credible alternative for data center AI workloads. Google, Amazon, and Microsoft are all developing custom silicon to reduce their dependence on Nvidia chips. Intel is attempting yet another datacenter GPU push with its Gaudi accelerators.

Against this backdrop, Nvidia’s reporting change is partly a defensive move. By isolating data center revenue as its own clean metric, the company makes it easier for analysts to track its market share in the segment that matters most. If competitors start chipping away at Nvidia’s dominance, the new structure will show it immediately, quarter by quarter, with no noise from other business lines muddying the picture.

That’s a confident move. You only make your most important number more visible when you believe it’s going to keep looking good.

Meanwhile, Nvidia is exploring financial models that could deepen its grip on the data center market even further. A potential $100B partnership with OpenAI involving a chip-leasing model for AI data center capacity would represent a fundamentally different relationship between chipmaker and customer. Instead of selling hardware and walking away, Nvidia would essentially become a landlord for AI compute, collecting recurring revenue as tenants scale their workloads.

That kind of arrangement, if it materializes, would make the data center segment even more central to Nvidia’s financial story. And it would make the new reporting structure look less like a cosmetic change and more like preparation for a business model shift.

What this means for investors

For anyone holding Nvidia stock or considering a position, the reporting change matters for a few practical reasons.

First, comparability. The old segments made it difficult to benchmark Nvidia’s data center performance against pure-play competitors or against hyperscaler capex budgets. The new structure fixes that. Analysts can now track Nvidia’s data center revenue directly against the capital expenditure disclosures from Microsoft, Google, Amazon, and Meta to gauge wallet share.

Second, transparency around edge computing creates a new metric to watch. If edge revenue starts growing at double-digit percentages quarter over quarter, it validates Nvidia’s diversification thesis. If it stagnates, it raises questions about whether the company can find its next growth engine before data center demand eventually normalizes.

Third, and this is the subtler point, the restructuring signals that Nvidia’s management sees the gaming and crypto-driven GPU cycles as less strategically relevant than they used to be. The company built its brand on gaming GPUs, and crypto mining demand created several boom-and-bust cycles that whipsawed the stock. By de-emphasizing those categories in its financial reporting, Nvidia is essentially asking investors to evaluate it as an enterprise infrastructure company, not a consumer hardware company with volatile demand drivers.

That reframing carries real valuation implications. Enterprise infrastructure companies typically trade at higher multiples than consumer hardware firms because their revenue is stickier and more predictable. If the market buys the new narrative, which it likely will given that 90% of revenue already comes from data centers, Nvidia’s valuation framework could shift accordingly.

The risk, of course, is that data center spending eventually hits a ceiling. Hyperscalers have been pouring hundreds of billions into AI infrastructure, but that pace of investment assumes AI workloads will keep scaling at current rates. If AI demand growth slows, or if cheaper alternatives to Nvidia silicon gain traction, having 90% of revenue concentrated in one segment becomes a vulnerability rather than a strength.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.