Is NVIDIA stock a good buy? NVIDIA (NASDAQ: NVDA) has become the dominant publicly listed beneficiary of the AI infrastructure expansion. Its latest reported quarter delivered more than $96 billion in revenue, approximately 75% gross margin and exceptional operating profitability. But at the $178.67 US closing price on 9 November 2026, the investment question is whether spending on AI accelerators and networking can keep increasing quickly enough to justify the company’s approximately $4.31 trillion equity value. Our evidence-based view is HOLD / SELECTIVE BUY ON WEAKNESS: the business is exceptionally strong, and the price is less demanding on forward earnings than its enormous market value suggests, but near-term volatility, concentrated customers and export restrictions make a strong-buy verdict premature.
This assessment is a comparison between a company and the price paid for it, not an assumption that strong growth guarantees strong shareholder returns. NVIDIA can continue winning AI infrastructure orders while its valuation multiple falls if customers moderate spending, competitors improve or supply grows faster than expected. Equally, the company could exceed conservative forecasts if more AI inference and software workloads make its systems economically indispensable.
Research cut-off: 9 November 2026, following the US regular-session close. Financial data reflect NVIDIA’s Q2 fiscal 2027 quarter ended 26 July 2026; the next quarter ended on 25 October but is scheduled to be reported on 17 November. Share prices, forecasts and scenarios are denominated in US dollars. Actual results may differ materially.
NVIDIA stock at a glance
| Market and financial measure | 9 November 2026 observation |
|---|---|
| NVDA closing price | $178.67 |
| Approximate outstanding shares | 24.147 billion |
| Implied market capitalisation | Approximately $4.31 trillion |
| Trailing twelve-month revenue | Approximately $302.97 billion |
| Trailing GAAP EPS | Approximately $7.91 |
| Trailing P/E | Approximately 22.6× |
| FY2027 adjusted EPS consensus | $9.31 |
| FY2028 adjusted EPS consensus | $15.92 |
| P/E against FY2027 adjusted EPS | 19.2× |
| Trailing free cash flow | Approximately $127.0 billion |
| Trailing FCF yield | Approximately 2.9% |
| Illustrative base DCF value | Approximately $200 per share |
| Research verdict | Hold / selective buy on weakness |
Why is NVIDIA stock falling despite exceptional AI earnings?
From the 29 October 2026 close near $213.66 to the 9 November close of $178.67, NVIDIA lost approximately 16.4%. The shares dropped 4.3% on 9 November alone, after heightened technology-market volatility. The observed 52-week range was approximately $164–$243. Sharp corrections are possible when a heavily owned AI stock meets a market suddenly questioning the economic returns from enormous data-centre spending.
This does not prove that NVIDIA’s customers have cancelled orders or that the company’s next quarter will disappoint. The market can reprice future expectations before the underlying revenue line weakens. The distinction matters because a share-price collapse and a collapse in corporate demand are different events. Investors should examine the historical market-price series separately from NVIDIA’s latest reported earnings.
A major near-term catalyst is already dated. NVIDIA announced that its Q3 FY2027 results would be released on 17 November 2026, with the conference call at 5 p.m. Eastern. As at the research cut-off, those results had not been published. This article deliberately does not treat an unreported quarter as fact. The official event announcement provides confirmation.
NVIDIA’s latest earnings: what does Q2 FY2027 show?
| Q2 metric | FY2027 Q2 | FY2026 Q2 | Year-on-year movement |
|---|---|---|---|
| Revenue | $96.221bn | $46.743bn | +106% |
| GAAP gross margin | 75.0% | 72.4% | +2.6 percentage points |
| GAAP operating income | $63.734bn | $28.440bn | +124% |
| GAAP operating margin | 66.2% | 60.8% | +5.4 percentage points |
| GAAP net income | $59.688bn | $26.422bn | +126% |
| GAAP diluted EPS | $2.46 | $1.08 | +128% |
| Non-GAAP diluted EPS | $2.22 | $1.01 | +120% |
| Data Centre revenue | $89.023bn | $41.096bn | +117% |
At this scale, doubling revenue in a year is extraordinary. Operating income rose even faster, reflecting pricing power, manufacturing leverage and the value of a tightly integrated GPU, networking and software platform. NVIDIA reported 75% GAAP gross margin and around 66% GAAP operating margin: levels that many hardware businesses never approach.
A necessary qualification is that GAAP net income includes gains on investments. NVIDIA reported $7.771 billion of net gains on equity securities in Q2 and $23.707 billion in the first fiscal half. These are not the same as gross margin on GPU sales or repeatable earnings from product operations. They explain in part why GAAP quarterly diluted EPS of $2.46 exceeded non-GAAP EPS of $2.22. The reconciliation deserves attention whenever an investment position affects earnings.
NVIDIA also revised its non-GAAP presentation: beginning in Q1 FY2027, stock-based compensation was no longer excluded from non-GAAP measures, and historical adjusted figures were updated. That improves consistency in one important respect, but investors should still review stock awards and share repurchases because reported EPS can benefit from a falling diluted share count. See the non-GAAP methodology and equity disclosures.
The Data Centre business: almost all of NVIDIA’s revenue
Data Centre revenue reached $89.023 billion, equivalent to approximately 92.5% of company-wide Q2 revenue. That concentration is a central characteristic of the investment. A diversified brand portfolio and global customer network do not change the fact that the near-term earnings engine is largely the global AI infrastructure build-out.
The remaining revenue came primarily from edge computing offerings, including gaming GPUs and other businesses. These operations are profitable strategic franchises in their own right, but they are relatively small beside AI data-centre demand. The sector’s extraordinary growth has therefore transformed what was once better known as a gaming graphics company into an infrastructure supplier for cloud computing and AI inference.
Hyperscale: powerful customers and difficult concentration
Hyperscale revenue reached $48.710 billion. The definition includes major cloud and technology customers purchasing large-scale AI and compute systems. Crucially, these are NVIDIA’s reported market platforms, not separate operating segments, and hyperscale plus AI Clouds, Industrial & Enterprise together constitute the $89.023 billion Data Centre total. Adding Data Centre again to the two subcategories would double-count revenue.
Customer concentration is measurable. NVIDIA’s filing identifies one direct customer representing 16% of Q2 revenue. Over the first six months, three direct customers represented 16%, 15% and 13% of consolidated revenue. Those percentages relate to parties purchasing from NVIDIA, such as distributors or cloud providers; they do not establish a named end user’s precise share. The distinction matters because the same underlying cloud operators may also purchase indirectly through manufacturing partners. The customer concentration note provides the definitions.
A key risk is that customers who currently need additional compute can eventually delay purchases as their previous hardware deployment matures. GPU orders are capital expenditure to the buyer, not recurring subscriptions. Microsoft’s own cash-flow analysis illustrates how rapidly hyperscaler data-centre spending can rise while free cash flow remains under pressure. NVIDIA benefits from this spending today but depends on customers ultimately making adequate returns from it.
AI clouds, industrial and enterprise: demand is broadening
The AI Clouds, Industrial & Enterprise category generated $40.313 billion, versus $16.928 billion a year earlier. The scale supports management’s view that demand extends beyond a handful of original foundation-model laboratories. Nevertheless, category membership, contractual arrangements and customer financing can change, and revenue alone does not prove that every AI customer has a sustainable profitable business model.
NVIDIA supports enterprise inference and training through compute systems, networking, CUDA and related software. Its advantage is a broad ecosystem: hardware performance, development tools, libraries, networking and integration make switching costs meaningful. But moat strength should be measured against alternatives, including specialised inference accelerators, hyperscaler-designed chips, AMD GPU systems, and software efficiency improvements that reduce compute needed per task.
What changes with Vera Rubin, Blackwell and AI agents?
NVIDIA described Vera Rubin as moving into production, with early production shipments commencing during the third fiscal quarter and systems running at partners. It also highlighted the Vera CPU, Spectrum-X networking and Groq 3 LPX inference hardware. These statements demonstrate product and partner activity, but not the future gross margin of the entire platform or the eventual number of paid deployments. The company product update and 10-Q discussion should be read together.
A new product transition can stimulate customer demand because faster chips deliver more useful compute per watt. The economic effect is more complex: greater performance can reduce hardware needed per unit of AI work, potentially increasing model usage while also pressuring pricing for older equipment. The direction of total demand depends on demand elasticity, capacity utilisation, customer profit margins and infrastructure constraints.
NVIDIA also competes by selling complete systems rather than just a discrete chip. High-speed interconnects, memory, cooling, racks and software are vital to GPU performance at large scale. This broadens its revenue opportunity but increases manufacturing and logistics complexity, dependence on suppliers, and risks from rapid architectural transitions.
Why CUDA and networking may matter more than raw GPU benchmarks
CUDA, specialised software libraries and the broader developer ecosystem have helped NVIDIA make its hardware practical to use across research and production. Customers invest substantial resources optimising their software on NVIDIA-based systems. Replacing the hardware can therefore involve extensive engineering, performance testing and retraining, not simply choosing a cheaper accelerator.
However, software ecosystems are not impregnable. Compiler improvements, open frameworks, alternative GPU architectures, application-specific chips and cloud abstraction layers can make it easier to shift workloads. A durable competitive advantage requires continual performance improvements and compelling total cost of ownership, not reliance on historical market share.
Networking and system-level performance are also critical. A cluster delivering faster individual GPUs can perform poorly when data cannot move efficiently between processors. NVIDIA’s integrated switching and interconnect offerings help its platform capture value beyond the compute die, but increase the cost of a full deployment to customers. Investors should monitor the usefulness and economics of complete deployments, not only chip specifications.
The biggest hidden issue: the difference between earnings and cash
NVIDIA’s first half of FY2027 generated approximately $74.421 billion of operating cash flow, compared with $42.779 billion in the corresponding prior year. Its expenditures for property, equipment and intangible assets were about $4.434 billion, with $92 million of related principal payments. Under NVIDIA’s stated FCF definition, that leaves approximately $69.895 billion of free cash flow in the six-month period.
The quarterly pattern is less smooth. Operating cash flow was $24.077 billion in Q2 versus approximately $50.344 billion in Q1. Quarterly free cash flow was about $21.341 billion in Q2. Changes in receivables, inventories and supplier payments can create large swings even when earnings continue rising. That is why comparing only annualised accounting net income with today’s market capitalisation can overlook working-capital risk.
| Six-month cash-flow item | H1 FY2027 | H1 FY2026 |
|---|---|---|
| Operating cash flow | $74.421bn | $42.779bn |
| Cash purchases of property/equipment/intangibles | $4.434bn | $3.122bn |
| Related principal payments | $0.092bn | $0.073bn |
| NVIDIA-defined free cash flow | $69.895bn | $39.584bn |
| Stock-based compensation | $3.954bn | $3.099bn |
| Change in receivables (cash-flow impact) | −$24.590bn | −$4.743bn |
| Change in inventories (cash-flow impact) | −$10.204bn | −$4.880bn |
The year-on-year free-cash-flow gain is real. Nonetheless, the first-half cash-flow statement also shows large cash uses from receivables and inventory growth. This reflects rapid scale, but the trend should be watched. A customer still owing money to NVIDIA cannot finance shareholder returns until it pays. Extended credit terms may also increase risk when AI customers are growing by borrowing or leasing heavily.
Using the latest four quarterly observations, NVIDIA’s trailing free cash flow is approximately $127.0 billion. At roughly $4.31 trillion in equity market value, that represents a 2.9% trailing cash-flow yield, or around 34 times trailing FCF. That is substantially less demanding than the market capitalisation sounds, but still depends on continued earnings growth. The next earnings report could alter the multiple significantly.
The balance sheet: liquid, but not free of financial commitments
| Financial item | 26 July 2026 |
|---|---|
| Cash and equivalents | $22.443bn |
| Marketable debt securities | $34.143bn |
| Cash plus marketable debt securities | $56.586bn |
| Short-term debt | $1.000bn |
| Long-term debt | $32.366bn |
| Simple net cash, excluding equity investments | $23.220bn |
| Marketable equity securities | $42.783bn |
| Non-marketable securities | $51.157bn |
| Accounts receivable | $63.059bn |
| Inventories | $31.575bn |
| Common shares outstanding | 24.147bn |
NVIDIA owns substantial equity investments in addition to its cash and debt securities. Those investments can produce large non-operating accounting gains and losses. A valuation should avoid treating all such securities as cash equivalents or assuming every dollar can be realised without tax, strategic costs or market impact. NVIDIA also increased borrowings as its investment portfolio and ecosystem commitments expanded.
NVIDIA repurchased approximately $19.7 billion of shares in Q2 and paid around $6.0 billion of dividends in the same period. In the first half it repurchased approximately $39.8 billion of common stock. Repurchases can improve future earnings per share if made at attractive prices, but they do not guarantee value creation when the market price is expensive. See the shareholders’ equity statement.
Is NVIDIA stock cheap or expensive at $179?
At the 9 November reference price of $178.67, approximately 24.147 billion shares imply a market capitalisation near $4.314 trillion. Trailing revenue was approximately $302.97 billion, making the trailing sales multiple roughly 14.2 times. Trailing GAAP EPS of about $7.91 implies 22.6 times earnings. These are high absolute expectations, but the growth rate is also unusually high.
Analyst estimates dated 9 November put FY2027 revenue near $411.8 billion and adjusted EPS at approximately $9.31. That means the share price is around 19.2 times FY2027 adjusted earnings. Analysts’ FY2028 adjusted EPS estimate of around $15.92 implies about 11.2 times FY2028 estimated earnings. These are conditional forward multiples: they do not mean investors can lock in those earnings today.
| FY2027 adjusted EPS | Illustrative P/E | Price implied by that assumption |
|---|---|---|
| $9.31 | 15× | $139.65 |
| $9.31 | 18× | $167.58 |
| $9.31 | 20× | $186.20 |
| $9.31 | 25× | $232.75 |
| $9.31 | 30× | $279.30 |
This exercise explains why seemingly cheap forward multiples can mislead. A 20-times multiple offers only modest upside if earnings meet the current FY2027 forecast, whereas 25 times produces a much larger target. If AI demand slows or the market begins discounting weaker profits in FY2028, the same earnings estimate could attract a much lower multiple. Valuation should therefore be anchored in sustainable longer-term cash flows, not a single fiscal year.
An independent ten-year discounted cash-flow valuation
Our base-case DCF begins with free cash flow of $190 billion in FY2027, consistent with the general scale of analysts’ current estimates. It then assumes annual free cash flow of $230bn, $270bn, $300bn, $325bn, $345bn, $365bn, $385bn, $405bn and $425bn through FY2036. These are our illustrative projections, not NVIDIA guidance. They require continued substantial AI infrastructure demand and increasingly efficient capital deployment by NVIDIA’s customers.
Discounting those cash flows at 9.0%, applying a 2.5% terminal growth rate, adding approximately $23.22 billion of simple net cash and dividing by 24.147 billion current shares yields a central indicative value of approximately $199.57 per share. The result is only about 12% above the $178.67 reference price, before allowing for uncertainty, future acquisitions and the possibility of additional share issuance or changes in capital commitments.
| Discount rate | 2.0% terminal growth | 2.5% terminal growth | 3.0% terminal growth |
|---|---|---|---|
| 8.0% | $225.15 | $238.49 | $254.50 |
| 9.0% | $190.66 | $199.57 | $209.96 |
| 10.0% | $164.91 | $171.13 | $178.24 |
| 11.0% | $144.97 | $149.47 | $154.53 |
The range is instructive. The base value changes from roughly $150 to $238 per share when the discount rate moves between 11% and 8% at 2.5% terminal growth. This degree of sensitivity should caution investors against calling any single DCF output the objectively correct price. AI market growth, supplier costs, software economics, customer concentration and capital expenditure cycles can materially alter free cash flow.
One caveat is especially important: the DCF treats only cash and marketable debt securities less financial debt as simple net cash. It does not separately add NVIDIA’s substantial holdings of publicly traded and private equity securities. Some of those investments are economically valuable, but adding their current carrying values without analysing potential obligations, strategic links or overlapping cash flows would risk overstating precision. NVIDIA’s earnings also contain equity-security gains that should not be confused with operating cash flow.
NVIDIA versus Microsoft, Micron and Intel
The AI supply chain has different economic models. NVIDIA supplies accelerated-computing platforms, networking and specialised software. Microsoft operates cloud and enterprise software services that monetise compute capacity. Micron supplies the HBM and DRAM memory critical to advanced processors. Intel designs CPUs and pursues an advanced manufacturing turnaround. A company can benefit from AI demand while facing completely different risks.
| Company | AI-related exposure | Primary risk in investment thesis |
|---|---|---|
| NVIDIA (NVDA) | AI GPUs, rack-scale systems, networking and software | Customer capex, export restrictions and competition |
| Microsoft (MSFT) | Azure compute, enterprise AI services and subscriptions | AI capex conversion into free cash flow |
| Micron (MU) | High-bandwidth memory and server DRAM/NAND | Memory pricing cycles and rival capacity |
| Intel (INTC) | Server CPUs, client chips and foundry production | Execution, foundry losses and dilution |
NVIDIA’s current margins and cash generation stand out, while Micron is more directly exposed to cyclical memory pricing. Microsoft’s cloud software has different revenue durability, but a major part of its infrastructure spending becomes sales for NVIDIA and its supply chain. Intel faces a much more uncertain manufacturing recovery. Investors considering semiconductor exposure should compare valuations, business models and overlapping sources of AI demand rather than simply buying all companies on the assumption that AI spending will always accelerate.
The greatest risks to NVIDIA shareholders
| Risk | How it could affect the stock | What to monitor |
|---|---|---|
| AI investment normalisation | Orders slow as customers digest prior infrastructure spending | Cloud-provider capex and utilisation |
| Customer concentration | A small number of buyers alters revenue forecasts | Direct and indirect customer exposure |
| Custom accelerators and AMD | Lower prices or market share | Real deployment costs and software switching |
| US-China restrictions | Lost addressable demand and tariffs | Export licences, H200 shipments and rule changes |
| Product-transition execution | Delays or lower yields increase system cost | Rubin shipments, Blackwell mix and margins |
| Financial-market repricing | Multiple compression despite profit growth | Forward EPS revisions and sector multiples |
| Capital allocation and investments | Non-operating volatility and strategic financing risk | Investment gains, losses and commitments |
| Working capital | Receivables and inventory absorb cash | Cash conversion, credit terms and stock levels |
Export restrictions are not a theoretical concern. NVIDIA said its H20 and H200 products were subject to licence requirements and significant practical restrictions involving China. The latest filing states that licensed H200 shipments represented less than 1% of Data Centre revenue, with certain products subject to US inspection and a 25% tariff that the company expected to absorb rather than pass on. Those restrictions constrain a material potential market and may encourage rival ecosystems. The 10-Q discussion details the exposure.
There is also a feedback-loop risk in AI financing. Infrastructure buyers may depend on external equity, debt or long-term customer commitments. If AI service revenue fails to grow fast enough, project financiers can reduce available capital, causing accelerator orders to slow. NVIDIA’s company earnings might remain strong for a time even if the broader AI-financing environment deteriorates. Our AI investment and recession analysis considers this wider macroeconomic relationship.
NVIDIA technical analysis and price levels
The price structure weakened materially in early November. NVDA closed around $213.66 on 29 October, $200.13 on 4 November, $181.60 on 5 November and $178.67 on 9 November. The most recent session traded between approximately $170.24 and $179.91. This is a sizeable pullback, not automatically a permanent change in the competitive position of NVIDIA’s technology.
The area around $170–$180 provides a recent trading reference, while $200–$214 marks an area where the price previously spent time before the decline. Neither range can be treated as guaranteed support or resistance. A more durable technical recovery would include higher lows, improving market breadth and an ability to hold above prior breakdown levels on meaningful volume. Conversely, further declines could coincide with more cautious forward earnings revisions.
Technical indicators are highly sensitive to lookback periods, corporate actions and data updates. Rather than invent precise RSI, MACD or 200-day averages, this analysis uses verified historical prices and points readers to the dated market-price data. Fundamental valuation should remain the main investment decision input.
NVDA stock forecast: bear, base and bull cases
The next three years could be unusually volatile. We assign subjective weights of 25% bear, 50% base and 25% bull to illustrate how different growth and multiple assumptions could affect the share price. The scenarios are not statistical probability estimates and are not Wall Street consensus. They do not incorporate future dividends or changes in outstanding share count.
| Scenario | Probability | 12-month price | 24-month price | 36-month price |
|---|---|---|---|---|
| Bear: AI capex slows, multiples fall | 25% | $125 | $100 | $90 |
| Base: growth persists, valuation moderates | 50% | $195 | $250 | $290 |
| Bull: AI expansion exceeds expectations | 25% | $285 | $350 | $430 |
| Probability-weighted | 100% | $200 | $237.50 | $275 |
The weighted 12-month outcome of $200 implies approximately 11.9% potential price appreciation. At 24 months, $237.50 implies about 32.9% cumulative upside. The $275 weighted 36-month outcome represents approximately 53.9% cumulative, or about 15.5% annualised, before dividends and dealing costs. The 36-month bear scenario implies a loss close to 50%; the bull case implies more than 140% upside.
The scenarios can be linked to earnings sensitivity rather than presented as arbitrary chart labels. For example, a 12-month base price of $195 would correspond to around 21 times FY2027 consensus adjusted EPS of $9.31. A weaker $125 outcome could reflect both downward estimate revisions and a lower earnings multiple. By year three, the implied per-share earnings required to sustain $290 depends on the future multiple investors will pay; at 20 times, that would require roughly $14.50 of annual adjusted EPS. These are examples of market assumptions, not actual company guidance.
Analysts’ average 12-month target is approximately $326.92 across 62 analysts in the 9 November data, much higher than the independent weighted outcome above. That target would imply around 83% price upside from the reference close if achieved. Analyst price targets are opinions that can adjust after earnings or market changes; their dispersion and assumptions should be reviewed rather than treating the average as fair value. Source: analyst consensus.
What could change NVIDIA’s investment outlook?
The immediately relevant catalyst is the 17 November 2026 earnings announcement. Management’s prior guidance called for Q3 revenue of $108.0 billion ± 2% and GAAP and non-GAAP gross margin of 74.0% ± 0.5 percentage points, excluding any assumed China Data Centre compute revenue. A reported result above guidance would not necessarily trigger a share-price increase if investors expected an even stronger number.
Beyond the quarter, key questions include the pace and economics of Vera Rubin adoption, whether more independent AI customers reach sustainable profitability, whether hyperscaler capital expenditure remains high, whether non-operating investment gains remain significant, and whether NVIDIA maintains cash conversion as receivables and inventory grow. These are measurable factors that can change an investor’s model more than headline claims about AI’s total addressable market.
A materially lower market price, without deterioration in the underlying earnings thesis, would also make the risk/reward more attractive. Conversely, a severe reduction in cloud customer spending or recurring cash generation would require a lower fair-value estimate. No technical entry level should be treated as a substitute for reassessing fundamental evidence.
Is NVIDIA stock a good long-term buy? Final verdict
NVIDIA is a highly profitable, strategically important AI infrastructure business, but its valuation is not risk-free at $178.67. The latest published results show revenue growth, enormous gross margins, exceptional cash generation and an integrated product ecosystem. On near-term consensus estimates, the shares look more reasonably valued than their multitrillion-dollar headline market capitalisation might suggest.
Our editorial conclusion is HOLD / SELECTIVE BUY ON WEAKNESS, rather than a categorical strong buy. Our central DCF assumption produces an indicative value around $200, giving only a modest margin of safety against a steep cyclical downturn or a lower valuation multiple. Investors who expect more prolonged high-margin AI infrastructure growth can justify larger valuations; they must explicitly accept the risk that their assumptions prove too optimistic.
The strongest counterargument to our caution is that AI inference, agents, robotics and software could generate far more compute demand than current customers’ plans imply, supporting NVIDIA’s earnings for longer than our forecasts. The strongest bearish argument is that customers may overinvest in capacity and introduce cheaper alternatives just as NVIDIA’s earnings base reaches an exceptional level. Both possibilities deserve disciplined attention.
NVIDIA stock FAQ
Is NVIDIA stock a good buy in November 2026?
Our assessment is Hold or selective buy on weakness at the $178.67 9 November close. NVIDIA is operating exceptionally well, but its reliance on AI data-centre spending and sensitivity to changes in earnings expectations make the risk/reward less certain than a simple forward P/E might suggest.
Why is NVIDIA’s share price falling?
Recent weakness reflects broad AI-sector valuation pressure and concerns about the future returns on infrastructure investment. A falling stock price is not itself evidence that quarterly orders collapsed. Investors should separate market sentiment from reported financial performance and check the next earnings release.
What is NVIDIA’s FY2027 stock valuation?
At $178.67, a FY2027 adjusted EPS consensus estimate of $9.31 implies a multiple of roughly 19.2 times. FY2028 estimated adjusted EPS near $15.92 implies about 11.2 times, but those estimates may change. A lower future multiple can offset substantial earnings growth.
How much of NVIDIA’s revenue comes from data centres?
Q2 FY2027 Data Centre revenue was $89.023 billion out of $96.221 billion total revenue, approximately 92.5%. The reported hyperscale and AI-cloud/enterprise subcategories together make up Data Centre. They should not be added on top of it.
When are NVIDIA’s next earnings?
NVIDIA scheduled Q3 FY2027 financial results for 17 November 2026, after the US market close, with its conference call at 5 p.m. Eastern. The official timing was confirmed before this article’s 9 November research cut-off. Results and guidance could materially change its valuation.
Is NVIDIA’s cash flow stronger than Microsoft’s?
NVIDIA’s cash generation is substantial, but the two companies have different business models and definitions of capital expenditure. NVIDIA primarily designs and sells accelerated computing systems, while Microsoft finances large cloud data-centre infrastructure. Compare each firm’s true after-capex cash flow and valuation rather than only absolute dollars.
Can AMD or custom AI chips threaten NVIDIA?
Yes. AMD, hyperscaler-designed accelerators and software-efficiency gains may reduce NVIDIA’s pricing power or share of future compute workloads. CUDA, system networking, integrated software and deployment scale are meaningful advantages, but no market position is guaranteed. Economics matter more than single benchmark results.
What would make NVIDIA stock a stronger buy?
A lower price relative to verified earnings, more durable cash generation, a diversified customer base or credible evidence that AI cloud providers are earning strong returns on their hardware investment would improve the investment case. Those factors can be examined after the next earnings report and subsequent guidance updates.
Research sources, calculations and limitations
Primary material: NVIDIA’s 26 August 2026 earnings announcement; NVIDIA’s July 2026 Form 10-Q; official 17 November earnings event notice. Secondary market data: price and trailing financial data; quarterly cash-flow history; dated consensus forecasts and analyst targets; market-price history. This analysis separates fiscal reporting periods, GAAP and non-GAAP metrics, company guidance, and independent assumptions.
The base DCF discounts explicitly stated annual free-cash-flow estimates over FY2027–FY2036 using 9.0% and 2.5% terminal growth, adds $23.22bn of cash and marketable debt securities net of interest-bearing debt, and divides by 24.147bn shares. It does not independently value all equity investments or model every potential contingent claim. Forecasts are sensitive to demand growth, margins, working capital, discount rates and customers’ capital spending. Scenario probabilities are subjective and not statistically validated.
Important notice: This article is general financial education and independent market research, not a personalised recommendation to buy or sell securities. Investment values and prices can decline substantially, including to zero. Forecasts may be wrong and market conditions can change rapidly. Readers should consider their own financial circumstances, objectives and capacity to bear losses and verify new disclosures before making investment decisions.
