Equity research

GF Securities: CPO/NPO Update

NVIDIA Spectrum-X CPO Scale-Out

Production Status: Spectrum-X Co-Packaged Optics (CPO) switches for scale-out networking have officially entered mass production.

Volume Forecasts:

2026E: 15,000 units

2027E: 100,000 units

Supply Chain Ramp: TSMC has expanded CPO inspection equipment capacity—specifically adding Insertion 2/3 capacity—alongside supply chain progress in Fiber Array Units (FAU), shuffle boxes, and system assembly.

Rubin Ultra Scale-Up Architecture & Optical Engines

Tray Design Options: Rubin Ultra is slated to adopt a 9-18-9 tray design, though mechanical challenges associated with the 0.75U height profile could drive a potential reversion to a 10-9-8 layout.

Design Impact: Architectural shifts in tray layout will have no impact on the optical engine (OE).

NVIDIA Platform OE Shipments:

2027E: 6 million units

2028E: 19 million units

Total Industry-Wide OE Shipments:

2027E: 11 million units

2028E: 40 million units

Amazon Trainium 4 & NPO Deployment

AWS Consumption Projections:

2H27E: 5 million OE units

2028E: 12 million OE units (primarily 6.4T specifications)

Configuration & Packaging: Trainium 4 is projected to feature 3 distinct configurations, with two configurations expected to adopt Near-Package Optics (NPO).

$Amazon(AMZN.US) $NVIDIA(NVDA.US)

WSJ: Nine top tech companies had some $3 trillion of off-balance-sheet commitments mostly related to AI.

A lot of spending going into this buildout...

$Meta Platforms(META.US) $Amazon(AMZN.US) $Microsoft(MSFT.US) $Alphabet(GOOGL.US)

J.P. Morgan: DRAM & NAND bit demand vs bit supply

DRAM bit supply/demand to be worse in 2027 before balancing in 2028, same for NAND.

Note: this is for bits. Does not mean shortage is solved in 2028.

$Micron Tech(MU.US) $EWY $DRAM $Micron Tech(MU.US)

Impressive...but at the same time hard to believe that by 2029, the market share for GPUs will be ~21% compared to ~79% for ASICs. Also implying a 17.5% market share for Nvidia lol.

I don't buy that a single bit.

$NVIDIA(NVDA.US) $AMD(AMD.US) $Intel(INTC.US) $Alphabet(GOOGL.US) $Amazon(AMZN.US) $Microsoft(MSFT.US)

Mizuho: InP & Memory

Supply Chain & Gating Factors

> InP Laser Supply: Remains tight into 2027E as multiple suppliers work to bring capacity online.

> NPO & CPO Dynamics: NPO continues to act as a near-term bridge for CPO, which is expected to ramp in 2H27/28E.

> CRDO Position: Believed to be the leader in Active Electrical Cables (AEC), with market ramping at a 98% 2024–29E CAGR, driven by additional upside from ZFO, ALC/MicroLED, and SiPho ramps.

> LITE Commentary: 1.6T ramps feature 200G accounting for >25% of revenues, with CPO delayed to 2H27. CRDO benefits from this alongside Neocloud strength for AEC.

Memory & Storage Pricing (NAND/DRAM)

> ASP Spikes: NAND/DRAM Average Selling Prices (ASPs) are projected up 406%/421% y/y in 2026E.

> Growth Moderation: Growth slows down in 2027E, with NAND expected to be flat and DRAM up 52% y/y.

AI DC Short-Reach Copper Cable Split

> Market Expansion: Total units grow steadily from 8M in 2024 to 23M by 2029.

> AEC Adoption: AEC units increase from 0M (2024) to 12M (2027) and 12M (2029), with total AEC penetration (including optical %) scaling to over 40% by 2029.

Estimated Wholesale 1.6T Cable Prices ($)

AEC: $225

ALC: $400

EML-based Transceiver: $700

CPO: $150

$Credo Tech(CRDO.US) $Lumentum(LITE.US) $Coherent Corp.(COHR.US) $Micron Tech(MU.US) $DDRAM $Sandisk(SNDK.US) $EWY

BofA: AI Compute

> Total Market Size: The total Data Center Systems TAM is projected to reach $2.2Tn in 2030E.

> AI vs. Non-AI Split:

AI Data Center Systems TAM accounts for the vast majority at $1.8Tn.

Non-AI DC Systems TAM accounts for $384bn (with Trad'l Cloud CPU representing ~$30bn of that value).

> AI Servers & Infrastructure Breakdown: Within the $1.8Tn AI Data Center TAM, AI Servers make up $1.4Tn, alongside $310bn for AI Networking and $85bn for Storage.

> AI Server Component Breakdown ($1.4Tn Total):

AI Accelerators: $1.1Tn

HBM (High-Bandwidth Memory): $277bn

AI CPUs: $180bn total value, which splits roughly 50/50 between AI Cluster Compute/Head Nodes ($90bn) and AI Agentic Standalone Nodes ($90bn).

Other: $53bn

$NVIDIA(NVDA.US) $AMD(AMD.US) $Alphabet - C(GOOG.US) $Amazon(AMZN.US) $Meta Platforms(META.US) $Dell Tech(DELL.US) $Super Micro Computer(SMCI.US)

Missed out on the space? No problem. Here is the recording.

We covered:

> Transition from copper to optical

> CPO scale-up timeline

> InP dependency on China

> The potential bottleneck in the InP supply chain

J.P. Morgan on memory LTAs: "A relatively misunderstood element within LTA is the resilience of pricing. Majority of the LTA discussion is currently centered around server memory. CSP and AI related memory demand is likely to be the majority (over 70% of total bit demand and 85+% of total revenue) of the contract. Server memory solutions carry a significant price premium over non-server applications and we believe the LTA mix increase will be pricing/margin accretive from a mid-to-long term horizon."

$Micron Tech(MU.US) $DRAM $EWY $Sandisk(SNDK.US) $SK Hynix(SKHY.US)

Memory Stanley: Memory

> Nature of Inflation: Shifted from viewing memory inflation as cyclical to recognizing it as structural.

> Demand Dynamics: Moved from the belief that higher prices cannibalize demand to realizing that higher prices and supply scarcity are actually accelerating demand.

> OEM Pricing Power: Changed from assuming OEMs absorb higher costs to seeing that OEMs are successfully passing through pricing and margin stacking.

> Supply Outlook: Replaced the expectation that supply would eventually catch up with the outlook that supply will remain constrained for years.

> Market Drivers: Expanded from a simple refresh cycle view to a combination of refresh, capacity expansion, and pull-forward demand.

> Overall Market Impact: Shifted from viewing memory inflation as universally bearish for hardware to understanding that memory inflation creates winners and losers.

$Dell Tech(DELL.US) $Hewlett Packard Enterprise(HPE.US) $Super Micro Computer(SMCI.US) $Micron Tech(MU.US) $EWY $DRAM $SK Hynix(SKHY.US)

J.P. Morgan: AI Accelerators

> Rapid Overall Market Expansion: Total annual AI accelerator unit shipments are projected to grow dramatically from 3.3 million units in 2023 to 23.3 million units by 2027E.

> NVIDIA's Continued Dominance: NVIDIA GPUs lead the market significantly throughout the period, growing from 1.7 million units in 2023 to an estimated 9.9 million units in 2027E, though their share of total units shifts as competition scales.

> Surge in Custom/Alternative Accelerators (ASIC/XPUs): Alternative hardware lines—including Google TPUs, AWS Trainium/Inferentia, Meta MTIA, and Microsoft MAIA—are scaling rapidly. By 2027E, custom ASICs/XPUs are projected to make up 53% of total unit shipments, surpassing traditional GPUs (47%).

> Significant Short-Term Growth Spikes: The market saw triple-digit year-over-year growth in 2024 at 128%, with strong continued expansion expected through 2026E (62%) and 2027E (43%).

$NVIDIA(NVDA.US) $Marvell Tech(MRVL.US) $Amazon(AMZN.US) $Alphabet(GOOGL.US) $Meta Platforms(META.US) $Microsoft(MSFT.US) $Broadcom(AVGO.US)

BofA: Nvidia's $500B Financing

> NVDA Underwrites Value, Not Debt: NVDA has signed Memorandums of Understanding (MOUs) with six top financiers (Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR) to mobilize over $500 billion in third-party capital via independent platforms. The financial burden sits with the consortium rather than NVDA's balance sheet.

> GPUs as an Investable Asset Class: With $500 billion of capital treating compute as an investable asset class, residual values must hold. NVDA supplies fungible and transferable compute across operators, and CUDA extends its useful life to keep resell/rental rates high and depreciation curves benign.

> Extending the AI Buildout Runway: Funding—rather than demand—has been the primary bottleneck. A $500 billion pool allows non-investment-grade buyers (such as labs, neoclouds, and sovereigns) to secure hardware at attractive rates, de-risking offtake and supporting long-term AI systems Total Addressable Market (TAM) paths.

> Key Debates and Risks Monitored: Notable cautions remain, including the fact that MOUs are not deployed capital (requiring real end-customers paying real money), potential power and regulatory pushback, input-cost inflation, and the opacity/complexity that new financing structures might introduce to AI buildouts.

Morgan Stanley: Semiconductor Sales

"Our CY26 memory forecast at $981bn is higher than our prior forecast and up meaningfully from $223bn last year. For CY27, we are assuming 29% growth y/y to $2.22 trillion, mostly due to memory dynamics."

$Micron Tech(MU.US) $DRAM $EWY $Sandisk(SNDK.US) $SK Hynix(SKHY.US)

Sumit Sadana, Chief Business Officer of Micron, at KeyBanc Technology Leadership Forum:

> "All of the efforts we are doing, we still don't have line of sight when supply is going to meet demand. Demand continues to escalate at a very rapid pace over time."

> "Very early stages of how this demand is going to proliferate."

> "HBM is growing significantly."

> "To produce 100 bits of HBM, we have to reduce 300 bits of DDR supply, 3:1 ratio. Closer to 4:1 for HBM4."

> "The number one constraint our customers tell us they have is DRAM."

> "We expect very robust business and financial performance for a very long time."

> "Aggregate demand across all of the segments is at much higher level than it has been."

> "Customers are telling us that despite the fact the prices are at a very high level, they are eager to get more supply because they are not able to meet their own business case requirements."

> "In data center, we are not able to meet any more than half of the demand our customers have."

> "Multi year signals on demand, not just 2026."

> "We have visibility on the engineering side that stretches out to roadmaps of our customers that are beyond 2030 timeframe."

> "Increased level of anxiety of our customers to secure supply."

> "We have signed the most number of SCAs across our industry."

> "Since our earnings, we have signed more SCA agreements with our customers."

> "These SCAs cover a long time horizon. Some customers that are smaller like automotive have 3 year SCAs. The SCAs that cover overwhelming amount of revenue under SCA is going to be 5 year type of terms through the end of 2030."

> "SCAs are binding commitments on purchases of volumes by year, by customer. Take or pay agreements with no contractual outs. Backed up by tremendous amounts of upfront cash like commitments."

> "SCAs are something LTAs never had. Some SCAs are going to be floating pricing, so pricing is going to be consistent with what the market prices at that time. Most of the volume under SCAs is going to have a ceiling price. The ceiling price at time of earnings with 16 SCAs is Q2 pricing. Since then, we have signed more SCAs. SCAs signed in the future is going to have ceiling pricing that is going to be consistent with whatever the market prices at the time the SCAs are signed in the future."

> "Some SCAs have no pricing mechanics and focus on market pricing."

> "The floor pricing is set at a level that provides a gross margin for us well above any prior peak in the cycles of the industry. These provide very high ROI."

> "SCAs have tremendous number of binding terms and no contractual way of backing out. More years can be added to the backend of the SCAs to extend them."

> "Huge amount of growth coming from robotics, which is going to be a ginormous growth driver."

> "This is not going to be the last time you see significant deficit of supply in this industry."

> "Physical AI in its infancy right now and has tremendous potential."

> "Lack of adequate data to train robots."

> "Each single humanoid robots is expected to have hundreds of GB of DRAM and TBs of SSD."

> "Customers struggling to get hands on adequate DRAM supply."

> "Availability of supply leading to despec among customers."

$Micron Tech(MU.US) $DRAM $EWY $Sandisk(SNDK.US) $SK Hynix(SKHY.US)

UBS: Micron & Memory

> BM Pricing & Premiums: Industry blended HBM ASP is up approximately 79% year-over-year, with HBM4E likely reaching >$30/GB.

> DRAM & NAND Contract Pricing:

For DDR, industry C3Q contract pricing (including LTAs) is modeled up 20% Q/Q, with MU specifically modeled up 20% Q/Q in C3Q and 11% Q/Q in C4Q.

For NAND, industry C3Q contract pricing is modeled up 20% Q/Q (down from 28% prior), and for MU specifically up 23% Q/Q (down from 35% prior).

Market Dynamics

> DRAM Tightness: HBM supply and demand have tightened beyond prior expectations, specifically for HBM4 and HBM4E, led by persistent supply tightness.

> HBM Consumption & Adjustments: C2027 HBM consumption is increasing from 58.7B Gb to 61.5B Gb, and MU HBM ASP is now modeled up 72% year-over-year (vs. +61% prior) with broadly unchanged bit shipments of ~11.65B Gb.

> NAND & SSD Strength: Stronger-than-expected server/storage SSD demand is offsetting weakness in PCs and smartphones, keeping the demand backdrop constructive.

> Competitor Impact: CXMT's DRAM bit supply share is expected to rise from ~7% in C2025 to ~9% in C2027, though capacity expansions are viewed as somewhat overstated due to yield and technology gaps. Meanwhile, YMTC is shifting incremental capacity toward DRAM over NAND, extending the NAND pricing cycle.

$Micron Tech(MU.US) $DRAM $EWY $SK Hynix(SKHY.US) $Sandisk(SNDK.US)

Nomura Securities: Global optical laser chip eco-system and competitive landscape

InP Wafer

AXT $AXT(AXTI.US)

Sumitomo Electric $5802.T

JX Advanced Metals $5016.T

Coherent $Coherent Corp.(COHR.US)

Lumentum $Lumentum(LITE.US)

Epi Wafer

IQE $IQE.L

Landmark Optoelectronics $3081.TW

VPEC $2455.TW

Coherent $Coherent Corp.(COHR.US)

Lumentum $Lumentum(LITE.US)

Sumitomo Electric $5802.T

Wafer Equipment

Aixtron $AIXA.DE

Veeco $Veeco Instruments(VECO.US)

Taiyo Nippon Sanso $4091.T

CW/EML Laser

Lumentum $Lumentum(LITE.US)

Coherent $Coherent Corp.(COHR.US)

Sumitomo Electric $5802.T

Mitsubishi Electric $6503.T

SiPho Foundry

Tower Semiconductor $Tower Semicon(TSEM.US)

GlobalFoundries $GlobalFoundries(GFS.US)

TSMC $Taiwan Semiconductor(TSM.US)

Chip Equipment

FormFactor $Formfactor(FORM.US)

Keysight Technologies $Keysight Tech(KEYS.US)

Anritsu $6754.T

UBS: Nvidia Vera Rubin BoM

Core Financial Breakdown (UBS BOM Analysis)

> Total Superchip Cost: $38,902 for the Vera Rubin core unit (1 Vera CPU + 2 Rubin GPUs).

> Total Memory Share: $24,297 (62%) goes entirely to memory semiconductors (HBM4 and SOCAMM2).

> CPU vs. GPU Memory Split:

SOCAMM2 (Vera CPU): $19,355 (49.8% of total cost) — a massive cost driver.

HBM4 (Rubin GPUs): $4,943 (12.7% of total cost).

> Memory Dominance in Costs: Memory semiconductors account for approximately 62% of the total cost of the Vera Rubin superchip. Out of a total cost of $38,902 per unit, memory components—specifically HBM4 and SOCAMM2—make up $24,297.

> Comparison to Previous Generations: In the previous Blackwell Ultra-based GB300, memory accounted for 53% of the cost; this has jumped to 62% for Vera Rubin. While the total cost of the superchip increased roughly 2.1 times compared to its predecessor, the cost specifically attributed to memory rose 2.5 times.

> Infrastructure Scale: The scale of the "Vera Rubin NVL72" system is immense. A single server rack contains 74.7 TB of DRAM. To put this in perspective, the LPDDR5X capacity in just one rack is roughly equivalent to the memory found in approximately 4,500 premium smartphones.

$NVIDIA(NVDA.US)

Third Bridge experts on CPO:

> Despite the hype, mass commercial adoption is not expected before the second half of 2027. Traditional pluggable optical modules are expected to remain the industry standard throughout 2026.

> CPO is likely to see early adoption in "scale-out" networking (connecting switches, GPUs, and CPUs) within distances under 500 meters. Deployment into GPUs themselves and "scale-up" chip-level systems is a harder engineering challenge, likely not seeing broad adoption until 2030–2032.

> Experts note that training workloads (which require extreme bandwidth) are better suited for CPO, while inference workloads are deterministic and remain well-supported by traditional pluggable modules.

> A primary hurdle for CPO is maintenance. With traditional pluggable modules, a failed component can be hot-swapped. With CPO, a failure in the optical subsystem might require replacing an entire board, leading to significantly higher costs and operational disruption.

> Nvidia and Broadcom are expected to dominate the CPO market, potentially controlling 70% of it by 2028. Nvidia is expected to capture around 50% market share through its leadership in AI infrastructure, while Broadcom could secure roughly 20% through continued expansion of its deployment footprint.

> Large cloud providers (like Google, Meta, AWS, Microsoft) are cautious about CPO because it risks "vendor lock-in." Pluggable modules allow these companies to mix and match components from various vendors, whereas CPO ecosystems are more restrictive.

> Through substantial investments in optical component companies like Coherent and Lumentum, Nvidia is positioning itself to control the entire AI value chain—from GPUs and networking switches to the optical modules themselves.

> Experts expect pluggable optical modules to remain the dominant solution for cloud giants such as Microsoft and Amazon throughout 2026. CPO adoption is expected to begin in earnest during 2027, reaching approximately 20% to 30% penetration by 2028.

Speedrun | Upcoming Events This Week - [August 10 - August 14, 2026]

Stay informed on events/news for the coming week.

> Iran Conflict & Strait of Hormuz:

President Trump has privately floated the idea of walking away from the Iran War without a nuclear deal if Iran fully reopens the Strait of Hormuz.

Negotiations have complicated as Iran is demanding billions in US payments and the removal of US troops.

Trump is expected to remain patient as long as gas prices stay stable, noting that Iran is likely unable to revive its nuclear work during his presidency.

> Memory vs. Optics Market Rotation:

Driven by a viral tweet from Jukan on optics versus memory, short-term rotation is favoring optics over memory.

Key catalysts include the death of Korean leveraged ETFs triggering LP redemptions, Nvidia shifting toward optics for Rubin Ultra clusters, and consensus forming that memory prices will peak within two quarters.

Analysts are beginning to pull back on aggressive DRAM ASP forecasts, creating selling pressure in memory while other tech/semis climb.

> Key Earnings & Economic Calendars

A heavy week for AI trade and tech earnings, featuring prominent names like Nebius, Super Micro Computer, Cisco, Coherent, Cerebras, Lumentum, CoreWeave, and Applied Materials.

> Economic Data Releases:

Wednesday: July CPI Inflation data.

Thursday: July PPI Inflation data.

Friday: July Retail Sales data and August Michigan

Consumer Sentiment data.

Link 👇

Earnings season is almost complete.

With nearly 90% of S&P 500 companies having reported, this earnings season is shaping up to be remarkably robust. Second-quarter earnings per share (EPS) are projected to surge 30% year-over-year when excluding investment mark-ups from Alphabet and Amazon—well above the 22% consensus forecast from July 1st. Factoring in those one-time gains for Alphabet and Amazon, EPS growth jumps to nearly 50% year-over-year. Furthermore, 76% of companies have surpassed EPS expectations, matching last quarter's beat rate, which marked the highest level seen since 2021.

$Alphabet(GOOGL.US) $Amazon(AMZN.US)

Tight MLCC supplies are driving a component scramble and price war, with Yageo reporting a surge in long-term customer supply contracts.

> Surging AI Demand & Tight Supply: Strong AI infrastructure requirements (servers, data centers) have driven delivery lead times out to 12 to 16 months, pushing major manufacturers like Yageo and Murata to near-full capacity.

> Bidding Frenzy & Price Premiums: Small and medium-sized buyers face compressed allocations and are paying 2x to 3x normal prices on urgent orders to secure scarce inventory, while tier-one customers take priority.

> Long-Term Contract Rush: AI clients are increasingly locking in multi-month or multi-year supply through binding long-term agreements with Yageo to mitigate future chain risks.

> Capacity Expansion: Yageo is actively scaling production sequentially across its global footprint (Kaohsiung, Suzhou, Vietnam, and Mexico) as new equipment arrives.

> Broad Industry Tailwinds: Leading Japanese producers are reporting similar momentum; Murata upgraded its annual profit forecast on data center demand, and Taiyo Yuden cited AI server growth accelerating the shift toward high-capacitance components.

Here is the Citi note on DRAM/NAND pricing.

Price Target & Valuation Adjustments

> Micron Technology (MU) Target Price: Trimmed to $1,150 (down from $1,400) based on an 8x P/E multiple (previously 10x) and a revised C27 EPS to reflect lower market multiples from mixed memory peer results.

> Rating: Maintained at Buy.

DRAM & NAND Pricing Outlook

> Deceleration: Both DRAM and NAND price increases are expected to decelerate quarter-over-quarter over the next four quarters, with prices projected to peak in 2Q of next year.

> Revised Pricing Assumptions:

DRAM Q/Q price adjustments shifted to 23%/9%/2%/0%, with 2H27 down 3% H/H.

NAND price assumptions moved to 29%/7%/0%/-1%, with 2H27 down 5% H/H.

FY27/28 EPS estimates lowered by 1%/2%.

Margins & Contracts

> Gross Margins: Expected to decline from current mid-80s levels and stabilize in the mid-70s next year as prices cool from a high base.

> Long-Term Agreements (LTAs): Micron has approximately 40% of DRAM bits locked under LTA pricing contracts.

Key Risks & Competition

> China Capacity Additions: Competition from Chinese manufacturers poses the biggest risk to the thesis:

YMTC: Plans to add 50k–60k wafer starts to its existing 200k operating capacity next year, boasts decent NAND tech, and aims to be the largest NAND player by 2030.

CXMT: Plans to add 50k to 350k DRAM capacity next year, currently has low yields, and targets ~600k by 2030.

While US restrictions limit direct sales into the US, European and other regional data center sales could indirectly impact Micron.

Mixed Memory Peer Performance

> Samsung: Reported memory sales and operating profit ahead of consensus driven by strong ASP growth in 2Q (mid-40% QoQ for DRAM, high-60% QoQ for NAND). Samsung is allocating 60–70% of planned capacity to LTAs, having signed 5 with major data center customers and finalizing 5 more.

> SK Hynix: Sales and operating profit came in just below consensus due to higher HBM exposure and slight product push-outs. DRAM ASPs rose 30% QoQ and NAND ASPs rose mid-50% QoQ. Signed 10 LTAs (typically lasting 5 years), including a $500B contract with Nvidia.

> Sandisk: Missed September-quarter revenue guidance due to more muted pricing, but secured 8 NBMs (diverse data centers and OEMs; weighted average duration of 4 years) covering 50%–66% of their bits in FY27/28 respectively.

$Micron Tech(MU.US) $DRAM $EWY $Sandisk(SNDK.US) $SK Hynix(SKHY.US)

Fubon Securities: Nvidia

Chip Output Forecast & Adjustments

> Rubin Ultra Adjustments: Following NVIDIA's cancellation or push-out of its Kyber rack and four-compute die Rubin Ultra GPU, the interposer size is now expected to see only a minor increase compared to Rubin.

> Interposer Estimates: The model estimates 12 interposers per CoWoS wafer for Rubin and 10 for Rubin Ultra.

> Total Chip Output: NVIDIA's total chip output is projected to increase to 12.4 million in 2027, up from 8.2 million in 2026.

> Short-Term Volume Revisions: Rubin's volume forecast for the short-term has been revised down to around 1 million due to earlier thermal design changes and production push-outs.

> Overall 2026 Output: Higher Blackwell volume in 3Q26 keeps the 2026 CoWoS chip output forecast largely unchanged.

> Supply Chain Impact: Production delays in Rubin negatively impact suppliers with higher dollar content; for example, KYEC revised its 2026 full-year guidance down to the high 30s from 40%.

BOM Cost & Pricing Estimates

> HBM4 Cost Increases: NVIDIA's HBM 4 cost is projected to rise to $31–32 per GB (up from $17–18 per GB for HBM 3e in 2026).

> GPU Pricing: NVIDIA's Rubin GPU price is expected to skyrocket to $78–80K, assuming NVIDIA maintains a gross margin (GM) of 75–80% on chips.

> ASIC & Other Projects: HBM costs for other GPUs and custom-ASIC projects are expected to reach $35–36 per GB, effectively doubling HBM costs by 2027.

$NVIDIA(NVDA.US) $Meta Platforms(META.US) $Microsoft(MSFT.US) $Alphabet(GOOGL.US) $Amazon(AMZN.US)

GF Securities: Nvidia's NPO Forecast

Nvidia NPO OE (Operating Environment/Units)

> Shipments: Projected to grow from 5.8 million units in 2027 to 8.1 million units in 2028.

> Growth: Represents a year-over-year (YoY) shipment growth of 39%.

Nvidia's TIA / Driver Financials

> ASP (Average Selling Price): Remains constant at $65 per OE across both years.

> Total Addressable Market (TAM): Expected to expand from $379 million in 2027 to $527 million in 2028.

> TAM Growth: Achieves a 39% YoY expansion.

Market Context & Breakdown (2028 Highlights)

> NVL72 Adoption: Expected to grow from 20% adoption in 2027 to 30% in 2028 (with 162k NPO NVSwitch trays/blades).

> NVL144 Adoption: Rises from 0% in 2027 to 15% in 2028 (contributing 63k NPO NVSwitch trays/blades).

> CSP (Cloud Service Provider) Scale-Up NPO: Demonstrates massive acceleration, with total market TAM for TIA/Driver surging by 482% YoY to reach $2,560 million in 2028.

$NVIDIA(NVDA.US)

BofA: Nvidia

Q2 Earnings Expectations

Q2 Sales Estimates: BofA expects a $3B–$4B beat, projecting $94B–$95B in sales (compared to the company's guidance of $91B).

Guidance Raise: Anticipates a 3%–4% raise for Q3 guidance to $107B–$108B, driven by the rollout of next-generation Vera Rubin shipments, new Vera CPU ramps, and robust cloud capital expenditure trends.

Memory Cost Inflation & Margins

Gross Margin Resilience: Despite rising memory pressures, NVIDIA's gross margins (GM) are expected to hold steady at a healthy ~73–74% over time (compared to ~75% currently).

Compute Racks Impact: The upcoming Vera Rubin generation will see a minimal ~60bps gross margin headwind on compute racks (total DRAM at ~12.7% of rack price) compared to Blackwell Ultra (~12.1%).

Offsets: Potential cost inflation is expected to be offset by NVIDIA's preferential sourcing through long-term agreements (LTAs), strong ties with memory vendors like SK Hynix, dynamic pricing power, and near-record GPU rental prices.

Ecosystem Partner Financing & Free Cash Flow

Manageable Commitments: Direct equity investments into ecosystem partners total roughly $70 billion to date (including $30 billion in OpenAI, up to $10 billion in Anthropic, and $5 billion in Safe Superintelligence).

Free Cash Flow (FCF) Coverage: This $70 billion commitment represents only ~15% of NVIDIA's projected $470 billion in FCF generation over the next two years (CY26–27E), leaving ample room to maintain its 50% FCF shareholder return pledge.

OpenAI/SB Energy Backstop: A potential $250 billion backstop for OpenAI's Ohio campus lease is a back-end-loaded (2028–30+) contingent guarantee, triggered only in the event of an OpenAI default.

GPU Rental Prices

Near-Peak Valuations: Rental prices for flagship NVIDIA GPUs (A100, H100, and B200) are holding firm near their historical highs as of August 2026.

Blackwell (B200) Leadership: The B200 commands the highest hourly rate in the index, trading steadily at $5.66/hr following a sharp upward re-rating earlier in the spring.

Hopper (H100) Stability: The H100 index sits at $2.80/hr, demonstrating steady upward momentum and persistent demand for mainstream AI training infrastructure.

Ampere (A100) Resilience: The legacy A100 remains locked near its multi-month peak at $1.64/hr, underscoring sustained utilization even for older-generation compute tiers.

$NVIDIA(NVDA.US)

SemiAnalysis: SpaceX 10GW

SpaceX's Gigawatt Ambitions & 2027 Projections

Massive Expansion: SpaceX aims to conservatively build and deliver an incremental 6–8 GW in 2027 alone, with potential to exceed 10 GW.

CapEx Requirements: At an estimated $50B per GW, this translates to $300B–$500B in CapEx for 2027, putting SpaceX on par with AWS and Google.

Path to $300B ARR: Driven by high-value, value-based pricing and rapid deployment timelines (3–5 months lead time), SpaceX is projected to reach $300B in ARR by the end of 2027.

Financing & Strategy: Funding is expected to be managed via Nvidia vendor financing (explaining Elon Musk’s focus on being Nvidia-exclusive) and industry-high pricing that pays back capital in under a year.

The Profitability of AI Inference

High Margins at Scale: Frontier model companies (OpenAI and Anthropic) can generate over $100B/GW/year in revenue when selling API inference on a GB300 cluster, significantly outstripping rental costs.

Revenue Climbs Up the AI Stack: On a GB300 cluster, annualized revenue per gigawatt scales dramatically from basic neocloud rental ($12B) up to $100B+ at the model-layer token API level.

Microsoft’s "2026 Awakening" & Largest Offtaker Role

Surge in Contracting: Microsoft has rebounded from a leasing pause to contract over 10 GW year-to-date across binding PPAs, ESAs, neocloud offtakes, and self-build construction starts, representing ~$300B in new commitments.

The Revenue Opportunity: Having reworked its OpenAI deal in April 2026 to remove the old 20% revenue share, Microsoft has a massive incentive to procure power rapidly to capture a $100M/MW/year revenue opportunity.

Azure Growth Impact: Contracted compute from partners like SpaceX could accelerate Microsoft Azure's revenue growth from ~42% to over 100% (peaking at 179% in 4Q27).

SpaceX’s Execution Speed

Proven Track Record: SpaceX and xAI have demonstrated unprecedented deployment speeds, such as building Colossus 1 (300MW) in 122 days and scaling the Southaven power plant from 27 turbines (~495MW) in Feb 2026 to 69 turbines (>1.2GW) by July 2026.

Onsite Power Generation: Hitting the 10 GW target heavily relies on scouting suitable land with easy permitting, leveraging fast greenfield/retrofitting methods, and utilizing massive onsite gas generation.