NVIDIA Q1 FY2027 Earnings: $81.6B Revenue, $75.2B Data Center, $80B Buyback, and Why the Stock Fell on a Beat

What You'll Learn

• What NVIDIA actually reported for Q1 FY2027 revenue, margins, earnings, and cash flow.
• Why Data Center remained the central engine while Edge Computing also grew.
• How the $91 billion Q2 outlook, China assumption, and product cadence affect the forward debate.
• Why a strong earnings beat can still produce a muted share-price reaction without proving the business is weakening.

NVIDIA Q1 FY2027 Earnings: The Report in One View

NVIDIA Q1 FY2027 earnings covered the quarter ended April 26, 2026 and were released on May 20. NVIDIA reported record revenue of $81.615 billion, up 20% from the previous quarter and 85% from a year earlier. GAAP diluted earnings per share were $2.39. The company’s non-GAAP diluted figure was $1.87.

The release also reported GAAP gross margin of 74.9%, non-GAAP gross margin of 75.0%, GAAP operating income of $53.536 billion, and GAAP net income of $58.321 billion. Free cash flow was $48.554 billion in the company’s investor presentation. These are realized Q1 results. They should be separated from the guidance and product claims discussed later.

The cleanest reading is not that every risk disappeared. It is that NVIDIA converted very large AI infrastructure demand into another quarter of high growth and high profitability. The harder question is what level of spending and product execution the current valuation already assumes.

Data Center Is Now the Earnings Center of Gravity

Data Center revenue was $75.246 billion, up 21% sequentially and 92% year over year. That figure represented more than nine-tenths of total quarterly revenue. NVIDIA said the result was driven by the ramp of Blackwell systems and demand for InfiniBand, Spectrum-X Ethernet, and NVLink solutions.

Under the previous sub-market presentation, Data Center compute revenue was $60.4 billion, up 77% year over year and 18% sequentially. Data Center networking revenue was $14.8 billion, up 199% year over year and 35% sequentially. The networking number is important because it shows that the AI infrastructure purchase is not only a GPU purchase. Customers also need switches, interconnects, systems, and software to operate large clusters.

NVIDIA also changed how it presents the business. It will report two market platforms, Data Center and Edge Computing. Within Data Center, Hyperscale includes public clouds and large consumer internet companies, while ACIE covers AI clouds, industrial, and enterprise markets. A reporting change does not create new revenue, but it can change how investors assess customer concentration and diversification.

Hyperscale, AI Clouds, and the Customer-Mix Question

NVIDIA’s CFO commentary reports Hyperscale revenue of $37.869 billion and AI Clouds, Industrial, and Enterprise revenue of $37.377 billion. Hyperscale was approximately half of Data Center revenue. The remaining share came from a broader group that included AI cloud providers, industrial customers, enterprise buyers, and sovereign customers.

That mix matters because investors are trying to judge whether AI infrastructure demand is broadening beyond a small number of hyperscalers. A diversified customer base can reduce reliance on one spending plan, but it does not remove common risks. Many customers still depend on the same cloud economics, electricity supply, model demand, and financing conditions.

The earnings call also described more than 80 partner data centers exceeding 10 megawatts and deployments across nearly 40 countries. Those are management disclosures. They help explain the commercial footprint, but they are not independent proof that every site will deliver a specific return on capital.

How to Read NVIDIA’s New Market-Platform Reporting

NVIDIA’s new presentation divides the business into Data Center and Edge Computing, then shows Hyperscale and ACIE inside Data Center. The change is meant to make the company’s growth drivers easier to follow. It also means readers should avoid comparing every new category directly with an old segment as if the definitions were identical.

The new framework is useful for understanding customer breadth, but it is still a reporting presentation. It does not change the consolidated revenue or create a new accounting segment by itself.

Edge Computing Adds a Second Growth Lane

Edge Computing revenue was $6.369 billion, up 10% sequentially and 29% year over year. NVIDIA defines this platform broadly. It includes PCs, game consoles, workstations, AI-RAN base stations, robotics, and automotive products. The company said strong Blackwell workstation demand was partly offset by slower consumer PC demand in an environment of higher memory and system prices.

This is strategically useful because it gives NVIDIA a business description beyond hyperscale data centers. It is also a smaller revenue pool than Data Center, so readers should not treat the Edge label as a substitute for Data Center growth. The financial weight remains concentrated in the infrastructure platform.

Physical AI and autonomous systems are longer-term opportunities. NVIDIA’s management described those markets positively, but opportunity statements should remain separate from quarterly revenue that has already been reported. The company still faces execution, customer adoption, supply, competition, and regulatory risks across these areas.

Margins, Cash Flow, and the Capital-Return Decision

GAAP gross margin was 74.9%, nearly flat sequentially and well above the 60.5% reported in Q1 FY26. NVIDIA’s CFO commentary attributes the year-over-year improvement mainly to lower inventory provisions, including the prior year’s $4.5 billion H20-related charge. That detail matters because a margin comparison can reflect both current operations and a difficult prior-year cost base.

Operating cash flow was $50.344 billion. Free cash flow, calculated by NVIDIA as operating cash flow less purchases related to property and equipment and intangible assets and principal payments, was $48.554 billion. The company returned approximately $20.0 billion to shareholders through repurchases and cash dividends during the quarter.

On May 18, the board approved an additional $80.0 billion share-repurchase authorization without expiration. NVIDIA also increased the quarterly cash dividend from $0.01 to $0.25 per share. The capital-return decision is material, but it does not by itself prove that the shares are undervalued. Investors still need to consider cash generation, reinvestment needs, dilution, valuation, and the durability of future demand.

Q2 FY27 Guidance Is Strong, With One Important Exclusion

NVIDIA guided to Q2 FY27 revenue of $91.0 billion plus or minus 2%. It guided to GAAP gross margin of 74.9% and non-GAAP gross margin of 75.0%, each with a range of plus or minus 50 basis points. It expects operating expenses of approximately $8.5 billion GAAP and $8.3 billion non-GAAP.

The most important qualification is that the outlook does not assume any Data Center compute revenue from China. That does not mean China demand is permanently absent or that policy risk is resolved. It means the company’s formal near-term outlook was constructed without that revenue assumption.

Guidance should be read as a management forecast under stated assumptions. The range is not a promise, and the absence of China Data Center compute revenue in the outlook is not a complete forecast of NVIDIA’s long-term China business.

Why China and Custom Silicon Matter to the Forward Debate

NVIDIA’s CFO commentary says no Data Center Hopper products were shipped to China during Q1 FY27, compared with $4.6 billion in the first quarter of fiscal 2026. The company also said its supply chain and customer decisions are subject to export controls, licensing, geopolitical conditions, and changing competition.

The 10-Q risk discussion adds a second issue. Some customers are developing custom application-specific integrated circuits and other products designed for particular workloads. Those designs may not need every feature of NVIDIA’s general-purpose data center systems. Customers may also offer cloud services that compete with NVIDIA’s AI cloud offerings.

This is not evidence that NVIDIA is losing its position. It is evidence that the market is not static. The company’s software, systems, networking, and developer ecosystem can support pricing and customer retention, while custom silicon can pressure selected workloads. Both points can be true at the same time.

Blackwell, Vera Rubin, and the Difference Between Revenue and Opportunity

NVIDIA’s earnings call said Vera Rubin was on track for production shipments in the second half of calendar 2026, starting in the third quarter. Management described a $1 trillion revenue outlook for Blackwell and Rubin from 2025 through calendar 2027, and said standalone Vera CPU sales were outside that particular forecast. NVIDIA’s CFO also described a $200 billion CPU total addressable market and nearly $20 billion of CPU revenue visibility for the year.

These statements are useful for understanding management’s strategy, but they are not realized revenue. A total addressable market is a market-size framing, not NVIDIA’s booked sales. A revenue outlook depends on product availability, customer orders, supply, competition, export rules, and the ability of customers to earn enough from their own AI services to keep investing.

For additional background on the AI investment cycle, see the site’s agentic AI finance guide and bond-market context guide.

Why a Beat Can Still Produce a Weak Share-Price Reaction

A strong earnings report is not evaluated in isolation. The stock price reflects expectations that existed before the release. A company can beat the quarter and raise guidance while still falling if investors expected an even larger beat, stronger margins, better China visibility, or clearer evidence that spending will remain durable into later years.

Reuters reported that NVIDIA shares fell 1.6% in extended trading after the results. Reuters linked the reaction to questions about competition and whether the AI buildout would remain durable into 2027 and 2028. CNBC described the report as strong while noting the post-earnings slide and concerns around custom chips, China, and high expectations.

That does not establish a single cause for every price move. The reaction can reflect positioning, options hedging, index flows, macro rates, or profit-taking as well as fundamental analysis. The practical lesson is that a beat is a data point. It is not a guarantee of a positive next-day return.

What Investors Should Measure After the Earnings Release

Readers assessing the next phase should track the same measures on a consistent basis. Revenue growth is important, but Data Center mix, networking growth, gross margin, operating expenses, cash conversion, and the Q2 delivery range show more about the operating model. China assumptions and export-control disclosures show where policy risk sits.

Product cadence also matters. Blackwell demand is a reported current driver. Vera Rubin is a forward product cycle. The further a claim sits from reported revenue, the more it should be treated as a scenario with execution risk. Management commentary can be informative without being a guarantee.

For related context, see the site’s rate-risk and AI market guide, the consumer-demand analysis, and the macro-growth context article.

Conclusion: A Strong Quarter Still Leaves a Harder Question

NVIDIA’s Q1 FY2027 report was objectively strong on the reported numbers. Revenue reached $81.615 billion, Data Center reached $75.246 billion, GAAP gross margin was 74.9%, and free cash flow was $48.554 billion. The board also expanded the buyback authorization and raised the dividend.

The market’s harder question is forward-looking. Can NVIDIA maintain the pace while customers build custom silicon, China revenue remains outside the Q2 outlook, and investors demand proof that AI infrastructure spending will stay durable? The report supports confidence in current execution. It does not remove the need to test future assumptions against future filings and guidance.

This is research and analysis only, not personalized financial advice. It is not a recommendation to buy, sell, hold, refinance, or change a portfolio. NVIDIA’s results, guidance, products, and market valuation can change, and personal decisions require current information and appropriate professional advice.