NVIDIA Q1 FY2027 Revenue Forecast: Analyst Consensus Heading Into May 2026

Heading into NVIDIA's (NVDA) fiscal first quarter of 2027, analyst consensus had coalesced around roughly $79 billion in revenue. That number represented a significant step up from prior quarters, but it was being set against a company that had recorded four consecutive quarters of accelerating revenue growth, a trajectory that made the $79 billion figure as much a baseline as a forecast.

The May 2026 earnings call would ultimately clarify how well that consensus tracked reality. The retrospective picture, supplied by NVIDIA's Q2 FY2027 results, makes the forecasting challenge concrete: Q2 FY2027 revenue came in at $96 billion, more than double year-over-year and the fourth straight quarter of accelerating growth.

Four Quarters of Accelerating Growth

The $79 billion Q1 estimate was constructed in an environment where every prior print had exceeded consensus. Analysts modeling Q1 FY2027 were weighing a business that had demonstrated sustained acceleration, with data center infrastructure buildout adding capacity on multiple continents simultaneously. Each quarter had not merely grown but grown faster than the one before it.

Q2 FY2027's $96 billion result, reported after Q1 closed, validated the direction of that consensus while indicating the pace of growth remained difficult to model precisely. The sequential and year-over-year patterns analysts observed heading into Q1 showed no sign of deceleration. That made the $79 billion estimate a reasonable calibration point but one subject to considerable upside uncertainty.

The Data Center Engine

Data center revenue in Q2 FY2027 reached $89 billion, up 18% sequentially. That figure comprised two distinct streams that analysts needed to weight separately. The hyperscale segment, covering revenue from large cloud providers, contributed $49 billion in Q2 FY2027, up 13% sequentially.

The other segment, ACIE, covering NeoClouds, enterprise customers, and sovereign AI programs, contributed $40 billion in Q2 FY2027, up 25% sequentially and 138% year-over-year. That $40 billion represented roughly half of total data center revenue, a proportion that had grown substantially in prior quarters and was itself a moving target for consensus models.

Sovereign AI as a Complicating Variable

The ACIE segment's growth rate was a significant modeling challenge heading into Q1 FY2027. Sovereign AI revenue alone tripled year-over-year as of Q2 FY2027. Sovereign AI customers, national infrastructure programs and government-backed compute deployments, were scaling orders quickly but on timelines that differed from hyperscale procurement cycles.

Amazon Web Services (AWS) illustrated the hyperscale side of the equation: the company committed to deploying an additional 2 million NVIDIA GPUs through fiscal 2029, alongside Vera CPUs and the upcoming Rubin GPU architecture. Multi-year procurement commitments of that scale anchored the hyperscale portion of consensus more reliably than sovereign AI volumes, which can shift between quarters based on procurement timelines and budget cycles.

For Q1 consensus modeling, the split between hyperscale and ACIE mattered because each carried a different forecast confidence interval. Analysts had relatively good visibility into hyperscale deployment timelines; sovereign AI procurement cadence was considerably harder to pin to a specific quarter.

Supply as the Binding Constraint

NVIDIA's guidance for Q3 FY2027, revenue of $108 billion, plus or minus 2%, made explicit what was already evident in the Q1 consensus debate: supply, not demand, was the binding variable. For full fiscal year 2028, NVIDIA projected approximately 70% revenue growth. That figure was explicitly described as supply-constrained; actual demand would support closer to 100% growth.

The gap between those two figures, roughly 30 percentage points of additional demand that cannot be met within the guided supply envelope, reframes how the $79 billion Q1 estimate is best read. Analysts setting consensus for Q1 FY2027 were not forecasting whether enterprise or hyperscale customers wanted to place orders. They were forecasting how much supply NVIDIA could bring to market and at what rate customers could absorb it.

This constraint meant that upside to Q1 consensus depended largely on whether chip packaging and memory supply chains had expanded faster than anticipated, not on demand signals. Demand was not the question. Delivery capacity was.

Gross Margin Dynamics

A secondary variable in Q1 FY2027 consensus models was gross margin. NVIDIA guided Q3 FY2027 gross margin to 74%. Looking further out, the company projected a trough of 71-72% in Q4 FY2027 before a recovery to 72-73% in the following fiscal year, driven by price increases on memory components taking effect across the supply chain.

The DRAM cost pressure creating that trough was visible to analysts constructing Q1 estimates. Higher memory costs, tied to the scale of Blackwell GPU production, were expected to compress margins during the ramp period. The Q1 FY2027 margin environment therefore sat in a transition zone where margin direction was known but the magnitude remained uncertain.

Analysts had to balance strong top-line momentum against a gross margin heading lower in the near term, which affected earnings-per-share estimates alongside revenue forecasts. The margin trough and the revenue ramp were running on overlapping timelines.

Per-Gigawatt Revenue Progression

The structural context behind the Q1 consensus range comes into sharper focus through NVIDIA's per-gigawatt revenue opportunity across successive GPU architectures. Each generation raises the revenue NVIDIA can extract from a given amount of data center power capacity: Hopper generates approximately $18 billion per gigawatt of data center capacity; Blackwell raises that figure to $25 billion per gigawatt; Vera Rubin lifts it further to $40 billion per gigawatt.

As data center operators moved from Hopper to Blackwell deployments, a transition already underway heading into Q1 FY2027, the same physical infrastructure generated more NVIDIA revenue per unit of power consumed. That architectural leverage embedded a structural upward bias in forward estimates. A cluster of the same megawatt capacity running Blackwell rather than Hopper represented roughly 39% more NVIDIA revenue.

The Blackwell ramp, combined with sovereign AI expansion and the AWS committed GPU deployment pipeline, gave analysts building Q1 FY2027 models a structural argument that the $79 billion consensus was a floor shaped by supply timing, not a ceiling defined by demand. The per-gigawatt progression through Vera Rubin extends that argument further into future fiscal years.

For a full breakdown of Q2 FY2027 results that provide retrospective context for the Q1 consensus, see the <a href="/podcasts/NVDA_Q2_2027">NVIDIA Q2 FY2027 earnings breakdown</a> or explore the full <a href="/groups/semiconductors">NVIDIA podcast feed</a>.

Key Numbers at a Glance

• Q2 FY2027 total revenue: $96 billion, more than double year-over-year
• Q2 FY2027 data center revenue: $89 billion, up 18% sequentially
• Hyperscale segment (Q2 FY2027): $49 billion, up 13% sequentially
• ACIE segment (Q2 FY2027): $40 billion, up 25% sequentially and 138% year-over-year
• Sovereign AI revenue (Q2 FY2027): tripled year-over-year
• Q3 FY2027 revenue guidance: $108 billion, plus or minus 2%
• FY2028 projected growth: approximately 70% (supply-constrained; demand implies closer to 100%)
• Q3 FY2027 gross margin guidance: 74%; Q4 FY2027 trough estimate: 71-72%
• Revenue per gigawatt: Hopper $18 billion, Blackwell $25 billion, Vera Rubin $40 billion
• AWS additional GPU deployment through FY2029: 2 million GPUs