Nvidia Q127: Everything everywhere all at once

Executive Summary

We upgraded our rating to BUY with a $280 price target (vs today’s price at $223). Q1 FY2027 was a beat-and-raise that validates rather than alters our constructive stance.

In our previous update, we highlighted our price target of USD 220 (vs USD 184 on 4^th March 2026), and the market has reacted extraordinary positive since then with the semiconductor bull run.

Read more: Nvidia Q426: The secret everyone knows

Everything everywhere all at once

Amid the AI megatrend, everything, everywhere, all at once, Nvidia remains the dominant platform provider across much of the AI infrastructure stack. No longer merely the maker of training GPUs, it is reaching into every arm of the build-out, a reach its new Hyperscale, Enterprise & Sovereign, and Edge segmentation now lays bare: spanning GPUs, networking, CPUs and emerging physical-AI applications.

Q1 FY2027 only deepened the moat, the next quarter guided higher still, while operating momentum has continued to exceed expectations. As the world races to build its thinking machines, Nvidia is increasingly monetising adjacent layers of the AI infrastructure stack beyond accelerators alone.

We remain positive.

We expect sequential revenue to continue building through FY2027 as the Vera Rubin platform ramps from the third quarter, gross margin to be sustained in the mid-70s, and free cash flow could exceed USD 200 bil this year, of which we expect roughly half to be returned via an expanded buyback and a higher dividend.

Figure 1: Our previous call for Nvidia in 2025 

Nvidia Q127 result

The quarter was a beat-and-raise that validates rather than alters our constructive stance. The consensus bull case on Nvidia, including secular AI-infrastructure demand, platform dominance and exceptional cash generation is well understood and, in our assessment, largely acknowledged by the market. The differentiated observation is therefore not that the business is strong, but that the equity continues to be priced for a deceleration the operating results are not yet exhibiting.

Table 1: Nvidia’s Q127 result

Source: Bloomberg Finance., iFAST Compilation, Data as of 21 May 2026.

Revenue of approximately USD 82 bil rose sequentially and beat the company's prior USD 78bn guide by a comfortable margin. Data Centre contributed approximately USD 75.2 bil, of which compute was roughly USD 60.4 bil, reflecting the continued ramp of Blackwell and GB300 and networking was approximately USD 14.8 bil, up nearly threefold y/y. The remaining ~USD 6–7 bil comprised Gaming, Professional Visualization, Automotive & Robotics and OEM. The composition of the beat is favourable for the thesis: it was concentrated in the core, higher-visibility business segments rather than in lower-quality or one-time lines. We would single out networking as an increasingly material disclosure; at this run-rate it is an increasingly meaningful growth contributor and reflects the company's transition from selling discrete accelerators to selling integrated, rack-scale systems.

Non-GAAP gross margin of 75.0% (74.9% GAAP) was flat sequentially and consistent with guidance, defying the bear expectation of systems-driven dilution. Operating expense continued to grow as the company funds its annual product cadence and widening software effort, but revenue growth outpaced it, preserving an operating margin in the mid-60s percent. This combination of resilient gross margin plus revenue-led operating leverage is the mechanism by which earnings compound faster than revenue, and it was clearly evident in the quarter.

Non-GAAP EPS of $1.87 beat the ~$1.77 consensus. GAAP EPS of $2.39 was inflated by an approximately USD 15.9 bil one-time mark-to-market gain on strategic investments; we treat this as non-recurring and exclude it entirely, modelling forward earnings on the operating run-rate. The non-GAAP effective tax rate was approximately 16%, and stock-based compensation ran at roughly USD 1.93 bil. We regard the underlying earnings quality as high, with the GAAP-to-non-GAAP gap this quarter being unusually wide and explicable.

Free cash flow of approximately USD 48.6 bil represented roughly 59% of revenue, underscoring best-in-class cash conversion. The balance sheet remains exceptionally strong, with net cash of approximately USD 72 bil even after substantial shareholder returns. Working-capital metrics bear monitoring, where days sales outstanding near 45 days and inventory days near 115, but in the context of guidance for an ~11% sequential revenue increase, we read the elevated supply position as a deliberate pre-build ahead of the Rubin ramp rather than as a demand-signal concern.

Figure 2: Quarterly revenue by segment (USD mil)

What has changed since our last update

Beyond the headline beat, several developments expand the medium-term opportunity set and the capital-return profile relative to the prior quarter:

• CPU monetisation quantified. Management framed the Vera CPU opportunity against a ~USD 200 bil TAM, with ~USD 20 bil of revenue this year, and confirmed Vera Rubin entering production in Q3 FY2027, extending the product cadence and the attach opportunity within each system sale.

Figure 3: Standalone Server CPU sales vs peers 

Source: Bloomberg Intelligence, iFASTCompilation, Data as of 21 May 2026.

• Enhanced capital return. The board authorised an incremental USD 80 bil of buyback and raised the dividend. Given free cash flow we estimate at >USD 200 bil in FY2027E, the company can return roughly half its cash generation while continuing to accrete net cash.

• Physical AI scaling. Robotics and autonomy surpassed USD 9 bil on a trailing-twelve-month basis, with a robotaxi roll-out targeted across multiple cities by 2028, which we believe remains only partially reflected in current valuation assumptions.

• Revised segmentation. Management re-cut its reporting into Hyperscale, Enterprise & Sovereign, and Edge categories, intended to evidence broadening demand beyond the largest cloud customers. Absent restated history, we continue to forecast on the legacy structure and treat the new disclosure as qualitative context.

Figure 4: Legacy quarterly revenue by segment (USD mil) 

• China assumption retained. The outlook continues to assume zero China data-center compute revenue. We regard this as appropriate conservatism that renders any policy normalisation incremental to an already-elevated guide.

Our Outlook

We are constructive across the forecast horizon. Our central scenario assumes the AI-infrastructure build-out remains in an expansionary phase through the medium term, with Nvidia retaining platform leadership even as the addressable market broadens beyond the largest cloud operators. The forward estimates underpinning our target are summarised in the table below.

Table 2: Revenue projection

Source: Bloomberg Finance L.P., iFAST Compilation, Data as of 21 May 2026.

Management's $91bn Q2 guide implies continued sequential acceleration, and we model full-year FY2027E revenue of USD 378 bil (+75% y/y), modestly above the ~USD 371 bil consensus, with non-GAAP EPS of $8.62 versus consensus of ~$8.44. Thereafter our base case turns deliberately cautious: we model FY2028E revenue of ~USD 499bn, broadly in line with consensus. We regard this as a conservative framing; should the China data-center channel partially reopen, or the build-out simply continue, there could be upside risk to estimates given the zero-China assumption embedded in guidance.

Data Center remains the engine, but our out-year assumptions are conservative. We forecast compute revenue scaling from ~$162bn in FY2026A to ~$388bn by FY2028E on the installed Blackwell base and the Rubin upgrade cycle (a massive yet relatively slower growth), while networking grows more modestly to ~$82bn. Outside the data centre, we model Gaming as a mature franchise in the high-teens, Professional Visualization growing steadily off a small base, and Automotive & Robotics building gradually from ~USD2bn toward ~USD 5bn as physical-AI and autonomy programmes scale.

Figure 5: Revenue forecast 

We assume gross margin sustained in the mid-70s (drifting from 75.0% to 74.0% as systems and networking mix grows) and operating expense deleveraging from ~9.0% of revenue toward ~8.3%, holding EBIT margins in the mid-60s percent and compounding non-GAAP EPS from $4.77 (FY2026A) to $11.54 (FY2028E).

Free cash flow of more than USD 200 bil in FY2027E scaling toward ~USD 272 bil by FY2028E funds a return of roughly half of FCF via the incremental USD 80 bil buyback and higher dividend, reducing the share count ~1.5–2% per year even after SBC dilution while the balance sheet continues to accrete net cash.

Underpinning all of this is the annual Blackwell, Vera Rubin, Feynman architecture cadence, the key driver of the revenue trajectory and of rising CPU and networking attach within each system, as illustrated in table below.

Table 3: Catalyst calendar

Source: Company announcement and estimates, Nvidia fiscal quarter end in late Apr/Jul/Oct/Jan.

The central debate: Durability vs Digestion

The investment case reduces to a single disagreement. The constructive view holds that the AI-infrastructure build-out is multi-year and multi-trillion-dollar in scale, management references global spending approaching USD 3–4 trillion annually by the end of the decade and that Nvidia’s hardware-software platform retains durable advantages that are difficult to displace. The cautious view holds that hyperscaler capital expenditure cannot compound indefinitely at the current pace, that a digestion period is likely, and that custom silicon from merchant vendors and the cloud operators themselves will erode Nvidia accelerator share over time.

We regard the custom-silicon and digestion risks as genuine, and our base case deliberately gives them weight: we assume a digestion plateau for the forward two year rather than extrapolating the current acceleration. The evidence in this quarter points to continued momentum, so this is a prudent rather than a pessimistic stance, and which, in our view, supports a favourable medium-term risk/reward profile.

Figure 6: GPU & ASIC unit forecast, 2023-2030e (as of 3^rd Dec 2025)

Risk

The principal downside risks are: 

1. A hyperscaler capital-expenditure digestion phase that slows revenue and compresses the multiple simultaneously — our bear scenario; 
2. Gross-margin compression, given the line's high sensitivity to EPS; 
3. Accelerating custom-silicon adoption eroding accelerator share; 
4. Escalation of U.S. export controls on advanced nodes or high-bandwidth memory; 
5. Supply constraints in advanced packaging or memory that cap upside. Upside risks centre on a China reopening (incremental to a guide that assumes zero), a faster physical-AI inflection, and a multiple re-rating toward the historical band on sustained beats.

The single most important monitorable is the sequential revenue trajectory. So long as each quarter continues to step up, we expect the constructive thesis to remain intact; a flat-to-down sequential print, absent an explicit supply constraint, would be the clearest signal that the digestion scenario is beginning and would prompt us to revisit both estimates and multiple.

Revised ‘BUY’, TP at $280. 29% upside potential in FY28

Our primary methodology values the equity on FY2028E non-GAAP EPS multiplied by a fair price/earnings multiple. We assign a base fair P/E of 24x, the SMH semiconductor-index multiple, valuing Nvidia in line with the sector rather than at its 35–45x historical premium, consistent with our conservative digestion base.

As such, we have computed a target price of $280 in FY28, a 26% upside potential from today’s price of $223.

Table 4: Nvidia’s valuation

Source: Bloomberg Finance L.P., iFAST Compilation, Data as of 21 May 2026.

Table 5: Income projection*

Source: Bloomberg Finance L.P., iFAST Compilation, Data as of 21 May 2026.
*Legacy revenue segmentation is being applied in our forecast

Disclaimers:

For specific disclosure, at the time of publication of this report, IFPL (via its connected and associated entities) and the analyst who produced this report hold a NIL position in the abovementioned securities.

This research report was prepared with the assistance of artificial intelligence (AI) tools. iFAST Financial Pte Ltd does not rely exclusively on AI for content generation; the content of this report – including all investment theses, ratings, price targets and conclusions – has been independently reviewed and verified by the research analyst(s) to ensure accuracy and professional integrity.