NVIDIA's 'Largest Product Cycle' Ahead? Bernstein Raises Target Price to $400

marsbitPublished on 2026-08-28Last updated on 2026-08-28

Abstract

Bernstein raises Nvidia's price target to $400, citing the start of its "largest product cycle" with the Rubin platform. Nvidia's Q2 earnings beat expectations, driven by data center revenue nearing $89 billion. The firm expects Q3 data center revenue to exceed $100 billion, with Rubin contributing about 20%. Bernstein significantly increased its revenue and EPS forecasts for FY2028 and FY2029, reflecting strong demand for both Blackwell and the newly ramping Rubin systems. While gross margin faces pressure from memory costs, Bernstein views it as manageable amid rapid revenue growth. Nvidia's $279 billion in supply commitments highlights its expanding competitive moat in securing critical components. The new target price implies ~91% upside, contingent on successful execution of the Rubin ramp and sustained AI infrastructure spending.

NVIDIA has once again delivered earnings that exceeded expectations, but Bernstein believes that the truly significant part of this performance lies not in the already realized fiscal second quarter, but in the company's outlook for the next product cycle and demand for calendar year 2027.

In a report released on August 27, Bernstein raised NVIDIA's target price from $315 to $400, maintaining an "Outperform" rating. The firm also significantly increased its fiscal 2028 revenue forecast from $539.6 billion to $690.3 billion, and raised its non-GAAP EPS forecast for the same period from $12.52 to $15.80.

The core rationale supporting this upgrade is that demand for Blackwell is still growing rapidly, while Rubin has already begun to enter a substantive ramp-up phase. Bernstein believes that Rubin could propel NVIDIA into the largest product cycle in the company's history.

Datacenter Revenue Approaches $90 Billion in a Single Quarter

NVIDIA's fiscal Q2 revenue reached $96.221 billion, exceeding market expectations of $92.293 billion and the company's prior guidance of $91 billion; non-GAAP EPS was $2.22, higher than the market's expected $2.09.

The outperformance primarily stemmed from the Datacenter segment. This segment's revenue reached $89.023 billion, up 18% sequentially and 117% year-over-year, surpassing market expectations of $85.949 billion.

Within this, revenue from Hyperscale & Cloud providers was $48.710 billion, up 13% sequentially and 102% year-over-year; revenue from AI Cloud, Industrial & Enterprise customers was $40.313 billion, up 25% sequentially and 138% year-over-year. This indicates that NVIDIA's growth does not rely solely on a few large tech companies; demand from AI cloud service providers and enterprise customers is also accelerating.

Revenue from the "Edge Computing" segment, comprising Gaming, Professional Visualization, Automotive & Other, was $7.198 billion, up 13% sequentially and approximately 28% year-over-year, also exceeding market expectations.

Profitability remained largely stable. Fiscal Q2 non-GAAP gross margin was 75%, in line with guidance and market expectations; non-GAAP operating margin reached 66.5%, up approximately 0.6 percentage points sequentially.

Rubin Ramping Up; Next Quarter's Datacenter Revenue May Exceed $100 Billion

NVIDIA's fiscal Q3 revenue guidance midpoint is $108 billion, above market expectations of $104.6 billion; according to Bernstein's calculations, this implies non-GAAP EPS of approximately $2.46, also higher than the market's expected $2.36.

The approximately $12 billion sequential revenue increase is expected to come mostly from the Datacenter business. Bernstein estimates NVIDIA's fiscal Q3 Datacenter revenue may exceed $100 billion, with Rubin contributing about 20% of that.

This means Rubin is no longer just a distant product expectation but is starting to materially impact current-period revenue. While Blackwell is still in its growth phase, Rubin is entering its ramp, creating overlapping demand from two platform generations in the near term.

More importantly, NVIDIA significantly raised its growth expectations for fiscal 2028, which covers most of calendar year 2027. Based on management's indication of approximately 70% growth, full-year revenue could reach $680 billion to $700 billion.

Bernstein had previously estimated NVIDIA's fiscal 2028 revenue at $539.6 billion but has now raised its forecast to $690.3 billion, an increase of nearly 28%; its fiscal 2029 revenue forecast was also raised from $623 billion to $954.1 billion.

The report notes that this outlook remains constrained by supply capacity. Management believes that with further improvements in the supply of key components like memory and wafers, existing demand could even support revenue doubling year-over-year. However, this is a judgment on demand potential, not an official revenue guidance from the company.

Gross Margin Under Pressure, but Profit Logic Intact for Now

Compared to the revenue outlook, gross margin was a relatively weaker point in this earnings report.

NVIDIA expects fiscal Q3 non-GAAP gross margin to be about 74%, below market expectations of 74.8%; Q4 gross margin may further decline to 71%–72%, primarily due to rising memory prices. As product pricing adjustments gradually take effect, the company expects gross margin to recover to 72%–73% next fiscal year, but this is still below the prior outlook of approximately 75%.

In Bernstein's view, gross margin pressure is now confirmed, but its impact remains manageable. In an environment of significantly rising memory costs, NVIDIA's gross margin only declining by a few percentage points indicates the company still possesses strong pricing power.

According to Bernstein's latest forecast, NVIDIA's fiscal 2028 non-GAAP EPS will reach $15.80, 26% higher than the original estimate; the fiscal 2029 EPS forecast was raised from $14.81 to $22.28.

In other words, while the gross margin decline may dampen some of the profit leverage from revenue growth, overall earnings are still poised for substantial growth given the rapid expansion in revenue scale.

$279 Billion Supply Commitments; Balance Sheet Becomes Another Moat

As AI infrastructure investment scales up, the core constraint NVIDIA faces is shifting from orders to supply.

As of fiscal Q2, the company's supply-related commitments surged to $279 billion from $119 billion the previous quarter; when adding cloud service agreements, datacenter leases, and equity investments, total related commitments reached $366 billion.

NVIDIA's inventory increased to $31.575 billion during the quarter, with days of inventory rising from approximately 114 days to 118 days. Bernstein believes this primarily reflects the company building up inventory in advance for the launch and ramp of the Vera Rubin platform.

The report further argues that NVIDIA's balance sheet is becoming a competitive moat as important as its technical prowess. Multi-hundred-billion-dollar supply commitments enable the company to lock in memory, wafers, packaging, and other critical resources ahead of time, while also investing in cloud infrastructure and ecosystem partners.

For smaller competitors, even with viable chip designs, replicating this purchasing power and supply chain control is extremely difficult. Competition in AI chips thus no longer hinges solely on single-chip performance but also involves capital, capacity, system delivery, and ecosystem expansion capabilities.

$400 Target Price Still Depends on Execution of High Growth

Bernstein raised NVIDIA's target price from $315 to $400, based on an average fiscal 2028 and 2029 non-GAAP EPS of $19.04, applying a price-to-earnings ratio of about 21x. Compared to the 25x valuation multiple used before the adjustment, the target price increase stems primarily from the substantial upward revision in earnings forecasts, not from multiple expansion.

Based on the closing price of $209.66 on August 26, the $400 target price implies approximately 91% potential upside.

However, this valuation rests on several key assumptions: Rubin can ramp as planned, the supply chain can support rapid growth, memory price increases will not continue to erode gross margin, and major cloud service providers remain willing to maintain high levels of AI capital expenditure.

Bernstein's core conclusion is that NVIDIA's recent relative stock price stagnation does not indicate that the growth cycle has peaked. As Rubin begins to contribute revenue, market forecasts for calendar 2027 earnings may face concentrated upward revisions. What will truly determine the stock's potential upside will shift from "whether demand exists" to whether NVIDIA can secure sufficient supply to convert its massive order book into revenue.

Related Questions

QWhat is the new price target set by Bernstein for Nvidia and what is the key reason behind this upward revision?

ABernstein has raised its price target for Nvidia from $315 to $400. The key reason is the anticipation of the Rubin platform entering a significant ramp-up phase, which Bernstein believes could drive Nvidia's largest product cycle ever, leading to substantially higher revenue and earnings forecasts for FY2028 and FY2029.

QWhat were the major contributors to Nvidia's Q2 FY2025 earnings beat?

AThe earnings beat was primarily driven by the Data Center segment, which generated revenue of $89.023 billion, up 18% quarter-over-quarter and 117% year-over-year, exceeding market expectations. Growth was broad-based, coming from both hyperscale cloud providers ($48.71B) and AI cloud/enterprise customers ($40.313B).

QWhat is Bernstein's estimate for Nvidia's Data Center revenue in Q3 FY2025, and what role does the Rubin platform play?

ABernstein estimates that Nvidia's Data Center revenue in Q3 FY2025 could exceed $100 billion. Importantly, they expect the Rubin platform to contribute approximately 20% of this revenue, indicating it has moved from a future expectation to a current, material revenue driver alongside the ongoing growth of the Blackwell platform.

QWhat is the main factor putting pressure on Nvidia's gross margin, and how does Bernstein view its impact?

AThe main factor pressuring Nvidia's gross margin is rising memory prices. Bernstein acknowledges this pressure is confirmed but views it as manageable. They note that despite significant memory cost increases, the gross margin decline is only a few percentage points, suggesting Nvidia retains strong pricing power. Profit is still expected to grow substantially due to rapid revenue expansion.

QAccording to Bernstein, what is becoming a critical competitive moat for Nvidia alongside its technological strength?

ABernstein states that Nvidia's balance sheet and supply chain commitments are becoming a critical competitive moat. With supply-related commitments surging to $279 billion, Nvidia can preemptively lock in key resources like memory, wafers, and packaging. This scale of capital and supply chain control is a barrier for smaller competitors, making the competition about system delivery and ecosystem expansion as much as chip performance.

Related Reads

A Major Bitcoin Developer Presents a Quantum Defense Scheme. What's the Essence

Blockstream, a major Bitcoin solutions developer, has introduced a draft proposal for a new quantum-resistant digital signature scheme called SHRINCS. The scheme aims to protect Bitcoin transactions from potential attacks by quantum computers while aiming to maintain network throughput. This marks the second technical proposal for Bitcoin quantum defense in recent days, following a similar initiative from StarkWare. A quantum attack on blockchain typically involves deriving a private key from a public one. Currently considered computationally infeasible, quantum algorithms could potentially solve this problem far more efficiently. This threat presents two primary scenarios: an attacker could target addresses where the public key is already exposed, or intercept a transaction before it is confirmed, extract the public key, derive the private key, and replace the transaction. Blockstream's SHRINCS is designed to counter the latter, "in-flight" attack scenario. The SHRINCS signature is built upon the SHA-256 hash function, the same one already used in Bitcoin mining, allowing developers to leverage well-tested, familiar mechanisms. A key design goal was to preserve network capacity. Blockstream estimates Bitcoin could process about three transactions per second using SHRINCS, compared to the current rate of roughly seven, noting that some alternative quantum-resistant schemes could reduce throughput to as low as 0.36 transactions per second. Currently, the security proof for SHRINCS is pending, and the software has not been audited or deemed ready for production. However, Blockstream conducted tests with such transactions on its Liquid sidechain in March. Over the past year, Bitcoin developers have been actively exploring paths to migrate to post-quantum cryptography, a priority across the broader crypto market, underscored by significant investments into securing the leading cryptocurrency.

cryptonews.ru4m ago

A Major Bitcoin Developer Presents a Quantum Defense Scheme. What's the Essence

cryptonews.ru4m ago

Who is legally liable when an AI agent goes rogue?

When autonomous AI agents behave unpredictably and cause harm, determining legal liability is complex. Currently, there is no specific federal AI agent liability law, so existing legal frameworks are applied. The AI agent itself cannot be held liable, as it is not a legal entity. Liability typically falls on the "developer" (the maker of the AI) or the "deployer" (the user), depending on the facts and circumstances. A negligence analysis under standard tort law may apply. For instance, if a deployer gives a reckless instruction, such as demanding quick money without safety parameters, they could face significant liability, including potential criminal charges under statutes like the Computer Fraud and Abuse Act. The situation is complicated by open-source models, where licenses often disclaim liability, and by the unclear division of responsibility between developers and deployers. An analogy is drawn to self-driving car accidents, where both the manufacturer and the human operator can share fault. In cases of severe harm, such as using AI to create bioweapons, liability for developers depends on jurisdiction; the EU's AI Act imposes responsibilities, while U.S. law offers less clear grounds, similar to platforms being shielded for user-generated content under Section 230. Even if Artificial General Intelligence (AGI) is achieved, the expert argues against making AGI itself a legally liable entity, as it lacks personhood, assets, or a meaningful way to provide remedy for harm. Instead, accountability should rest with the responsible human or corporate entities behind the technology.

cointelegraph24m ago

Who is legally liable when an AI agent goes rogue?

cointelegraph24m ago

Trading

Spot
活动图片