Nvidia Corp. (NVDA) Report Interpretation

Overview

This pre-earnings report argues that Nvidia is positioned for a solid 2Q and guidance upside, but that the earnings call must address several issues central to the next phase of AI-infrastructure growth: the scope of customer financing, the pace of Rubin adoption, gross-margin pressure, CPU demand, and competition. Goldman Sachs reiterates Buy and its $285 target.

Core views

Goldman Sachs expects Nvidia to report a solid fiscal 2Q and to issue fiscal 3Q guidance well above Street expectations. Its forecasts assume 2Q revenue of $93.05bn and EPS excluding stock-based compensation of $2.18, versus Street estimates of $92.01bn and $2.05. For 3Q, it estimates revenue of $110.66bn and EPS of $2.63, compared with consensus of $103.99bn and $2.36. The anticipated upside is driven principally by Datacenter: Goldman Sachs forecasts $86.53bn for 2Q and $103.84bn for 3Q, respectively 1% and 7% above consensus. Tight GPU supply-demand conditions, positive 2026 capex revisions from US hyperscalers, and commentary on near-term GPU capacity shortages support this view. However, the institution considers the setup demanding because the stock had risen 12% in two weeks. The first major earnings-call issue is Nvidia’s recently announced $500bn financing platform with financial partners, intended to support AI-infrastructure buildouts by AI labs, enterprises, and AI clouds at attractive rates. Goldman Sachs notes that details remain limited and that company commentary suggests Nvidia could provide residual-value support for up to 25% of an opportunity, assessed project by project. Investors will seek clarification on the maximum capital commitment and on whether such commitments could constrain Nvidia’s ability to sustain its stated buyback and dividend objectives in a downside case. Goldman Sachs views measured capital outlays and continued strong capital returns as evidence that could support further re-rating. A second focus is the shape of the Rubin product ramp in 3Q and 2027, including whether adoption broadens beyond hyperscalers to AI labs, neoclouds, and sovereign customers. The report highlights any quantitative update to Nvidia’s cumulative $1 trillion revenue guidance for Blackwell, Blackwell Ultra, and Rubin products over CY2025-27 as a key indicator of Rubin’s trajectory and potential stock impact. Goldman Sachs expects the Rubin ramp to begin contributing in 3Q, while its longer-term estimates remain above consensus: FY2027E revenue of $410.85bn and EPS of $9.50 are 4% and 6% above Street, rising to FY2028E revenue of $635.12bn and EPS of $15.25, 13% and 19% above consensus. Margin durability is another central issue. Nvidia reported gross margins above 75% in each of the past two quarters, while input-cost pressure—especially in HBM memory—is rising. The institution expects management to frame margins over the coming year in the context of performance and cost-per-token economics across product generations. Its checks suggest Nvidia is seeking to reduce HBM content in some Rubin-based servers by up to 50%, which is relevant to how the company may manage input costs. Goldman Sachs forecasts gross margin including stock-based compensation of 75.0% in both 2Q and 3Q, above consensus by 80 basis points and 16 basis points, respectively. The report also flags potential CPU demand from agentic AI. Nvidia’s CPU-only racks are expected to begin shipping in 2H26, and management commentary on adoption of agentic AI, CPU-based rack-level systems, and the medium-term size of that opportunity will be important. After results, Goldman Sachs expects the market to focus on hyperscalers’ monetization and ROI, agentic-AI uptake across consumer and enterprise workloads, execution by neocloud and other non-hyperscaler customers, competition from custom XPUs and AMD solutions, and gross-margin effects from memory and other input costs. Goldman Sachs reiterates Buy and a $285 12-month target, based on a 30x P/E multiple applied to normalized EPS of $9.50. It states that NVDA trades at 14.4x its CY2027E EPS estimate, which it views as a wide discount to fair value, and notes that its CY2026 and CY2027 estimates are 6% and 19% above Street expectations. The stated downside risks are slower AI-infrastructure spending, share erosion from greater competition, margin erosion from competition, and supply constraints.

Analysis framework

Goldman Sachs compares its quarterly and multi-year revenue, margin, operating-income, EPS, and segment forecasts with Street consensus, then connects the expected Datacenter upside to GPU supply-demand conditions and the Rubin ramp. It frames the call around financing, capital allocation, product adoption, costs, and competition, and values Nvidia using a normalized-EPS P/E multiple.

Methodology notes

• Valuation methodsP/E and PEG Valuation

  Price-to-earnings valuation

  Goldman Sachs derives its $285 target by applying a 30x P/E multiple to normalized EPS of $9.50.

• Industry AnalysisSupply-demand framework

  GPU supply-demand analysis

  The report links tight GPU capacity, hyperscaler capex revisions, and demand for AI infrastructure to expected Datacenter revenue and guidance upside.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Nvidia Corp. (NVDA)

  Primary covered company; expected Datacenter-led earnings and guidance upside from tight GPU supply-demand conditions and the initial Rubin ramp.

  Strengths

  Tight GPU supply-demand trends, expected Datacenter upside, a broad AI-infrastructure product cycle, and potential CPU demand from agentic AI.

  Weaknesses

  Elevated expectations after a 12% two-week share-price move and uncertainty around the scale of customer-financing commitments.

  Comparison

  Goldman Sachs forecasts FY2027E EPS of $9.50 versus Street consensus of $8.98 and FY2028E EPS of $15.25 versus $12.85.

  Risks

  Slower AI-infrastructure spending, competitive share erosion, margin erosion, and supply constraints.

• Advanced Micro Devices

  Competitor referenced through its MI450X Helios rack-scale systems and expanding customer set of AI model builders.

  Comparison

  Its AI-system launch is cited as part of the broadening competitive environment for Nvidia.

Key data

• 2Q revenue estimate$93.05bn1% above Street consensus of $92.01bn
• 2Q EPS estimate$2.18Excluding SBC; 6% above Street estimate of $2.05
• 3Q revenue estimate$110.66bn6% above Street consensus of $103.99bn
• 3Q EPS estimate$2.63Excluding SBC; 12% above Street estimate of $2.36
• FY2027E revenue and EPS$410.85bn and $9.504% and 6% above Street, respectively
• FY2028E revenue and EPS$635.12bn and $15.2513% and 19% above Street, respectively
• Partner financing platform$500bnAnnounced to support AI-infrastructure buildouts
• Target valuation$28530x P/E applied to normalized EPS of $9.50; 31.0% upside from $217.55

Impact & implications

The report argues that evidence of sustained hyperscaler spending, disciplined financing-related capital commitments, continued capital returns, and a credible Rubin ramp could support further re-rating. Conversely, the same issues may determine whether elevated expectations can be met, particularly if financing obligations, input costs, or competitive pressure weigh on margins or market share.

Risks

• AI-infrastructure spending could slow.
• Nvidia could lose share as competitive intensity increases.
• Increased competition could erode margins.
• Supply constraints could limit performance.

What to watch

• Details of Nvidia’s maximum capital commitment and residual-value support within the $500bn financing platform.
• Whether financing commitments affect stated buyback and dividend targets in a downside scenario.
• The timing and customer breadth of the Rubin ramp in 3Q and 2027.
• Any update to cumulative CY2025-27 revenue guidance of $1 trillion for Blackwell, Blackwell Ultra, and Rubin.
• Gross-margin guidance amid HBM and other input-cost pressure.
• CPU-only rack shipments in 2H26 and adoption of agentic-AI-driven CPU systems.
• Hyperscaler monetization, neocloud execution, and competitive pressure from custom XPUs and AMD.