Nebius: Mega Microsoft Deal, 5x Growth in Power Fueling AI Cloud Hypergrowth, But High Risk Remains

Posted on September 12, 2025June 30, 2026 by io-fund

The trend toward neoclouds is a high risk/high reward opportunity for investors. Nebius is similar to CoreWeave, dubbing itself as AI native cloud infrastructure, which means the infrastructure was built specifically for AI workloads with architectures built on bare metal servers instead of hypervisor layers, for example. As pointed out in our CoreWeave analysis, this along with a few other optimizations can result in significantly faster training.

Nebius offers an Nvidia-optimized cloud platform for teams that need to adjust compute resources, want high-performance storage and an easy-to-use AI environment yet do not want to manage the infrastructure or AI operations.

The stock surged earlier this month off the announcement of a mega deal with Microsoft worth $19.4 billion. This deal signals just how supply constrained the market is, given Microsoft is willing to partner with a neocloud to scale quickly.

Even with the big moves in neocloud stocks we’ve seen recently, they remain high risk and they are not in the “quality” bucket given their financials are messy. Nebius is even more complicated than CoreWeave given they also own an autonomous driving division, among other investments, and was formally the company Yandex. The Russian-owned Yandex was a high-profile internet search company (the Google or Baidu of Russia), which saw its stock halted after the Russian invasion of Ukraine.

Given Nebius’ roots are from Yandex, one might question if Nebius is truly AI-native as some of the Finnish data center in Europe was built in 2014. However, that is also where one of Nebius’ strength lies, which is the company offers European data centers which helps a company like Microsoft to expand quickly overseas.

To serve the AI-native market, Nebius has been building out colocation sites to increase capacity and lower latency for the incoming inference market. Companies like Cloudflare and Shopify are early customers, both of which need to power inference at the edge. The company is also expanding beyond Europe with data center expansions in New Jersey and Kansas.

With that said, Nebius is a high-risk stock given its success depends on how much capital the company can raise, and the likelihood it remains in CoreWeave’s shadow is high. Regardless of how Nebius competes with CoreWeave, it remains an AI bubble stock as the company has to hope the stock prices goes up to raise cash, which will dilute shareholders (or raise debt). It’s a vicious cycle as during any months/quarters that AI stocks are soft, Nebius stock will carry outsized execution risk.

Ultimately, we are stepping away from the stock. While it’s clear that Nebius could do well in an AI-driven market and the stock could extend rapidly, we want to be prudent about what will hold up should we see a softer AI narrative. We like other names in our portfolio better in terms of participating in the upside while protecting to the downside for when we do get the inevitable tech selloff.

Up to $19.4 Billion Deal with Microsoft, More to Come?

Last week, Nebius signed a mega deal with Microsoft worth up to $19.4 billion through 2031, for capacity at its new AI data center in New Jersey. The announcement sent Nebius shares up 50%, with the contract value being worth more than the company’s $15.5 billion market cap at the time of announcement.

Nebius outlined that it plans to bring GPUs online in several tranches, with initial capacity later in 2025 and ramping throughout 2026. At present, the deal is worth $17.4 billion, with Microsoft having the ability to sign a $2 billion extension should it determine a need for additional GPU services or capacity. Nebius expects the deal will help it greatly accelerate AI cloud growth in 2026 and beyond.

The deal does have a few major contingencies, mainly that Microsoft has the ability to cancel the contract if Nebius fails to meet agreed delivery dates and if it cannot provide alternative capacity to make up for the delay. It’s likely that Nebius will prioritize the development of the New Jersey data center above all else in the near-term to ensure it remains in good standing with Microsoft as the deal offers immense revenue upside.

Unlike miners, who already have pre-built infrastructure ready for retrofitting, this is a build-to-suit development, meaning Nebius is financing the buildout of the infrastructure, racks and related equipment and partnering with a developer who owns the land and power (in this case DataOne).

Nebius expects to fund the capex for this data center with a combination of cash flow from the deal and debt secured against the contract, while noting that additional financing options may be pursued to “enable significantly faster growth than originally planned.” The latter has already panned out, with Nebius raising $3.75 billion in a combined debt & share sale. This is important as the terms of the deal do not indicate upfront prepay, with deferred revenue only $19.3 million, suggesting Nebius will likely have to deliver the first tranche of power before payments commence.

Breaking this down, the $17.4 billion deal is worth approximately $3.48 billion in average annual revenue over the five-year period, or scaling to the several billion in revenue by 2031. Overall revenue estimates for 2028 through 2030 have been raised by $2 billion to $4 billion following the announcement, implying the ramp will be felt primarily in the later stages of the deal.

While the multi-billion dollar upside is certainly a positive, CEO and founder Arkady Volozh hinted that subsequent deals like this are possible: “In addition to our core business, we expect to secure significant long-term committed contracts with leading AI labs and big tech companies. I’m happy to announce the first of these contracts, and I believe there are more to come.”

Considering the New Jersey data center has 300MW capacity with potential expansion to 400MW, Nebius theoretically can support more deals of this size with 1GW+ in its pipeline and aggressive growth plans.

Targeting 5x Growth in Power by End of 2026

Nebius is aggressively ramping up its capacity and has outlined a plan to increase its connected power by ~5x by the end of next year, though it is still targeting 100MW of active power by the end of this year.

This quarter, Nebius stated that it is aiming to have 220MW of connected power by the end of the year, which they define as either currently active or that can be activated upon GPU installation. Management maintained its guidance for >100MW of this 220MW to be active by year-end, though this may be revised higher to account for expedited expansion for Microsoft’s deal.

This includes recent expansion at multiple data center campuses and new campuses soon to be operational:

In Finland, Nebius is currently tripling its capacity to 75MW, which can host up to 60K GPUs.
In New Jersey, Nebius has 300MW under construction, with the first ~100MW to be connected this year; some reports suggest the site can expand to 400MW. Considering this is the site powering the Microsoft deal, it’s likely Nebius will prioritize power connections on an accelerated rate here.
Nebius’ Kansas City, Missouri data center went live in Q1 2025, featuring primarily H200 GPUs, and will deploy Blackwell GPUs through year end. The first phase can expand up to 40MW and host 35K GPUs.
Nebius is launching its first UK data center in Q4 2025, expected to feature 4,000 Blackwell Ultra GPUs.
Nebius is also launching a new site in Israel, joining its global footprint with other operational data centers in Paris and Iceland.

Overall, Nebius is in the process of securing more than 1GW of power by year-end 2026, or nearly 5x growth from 2025’s current guidance for connected power. This includes two new greenfield sites in the US where Nebius is in advanced discussions; management prefers greenfield construction as they can achieve TCOs around 20% below market averages by controlling building design, power flow, and server design and installation.

This growth in power supports Nebius’ medium-term expectations for scaling its fleet, from ~20K in 2024 to 60K in 2025 to 240K over the next few years. This paves the way for Nebius to continue smoothly along its hypergrowth phase, progress towards its medium-term revenue target of ‘several billions’ and support more hyperscaler deals.

To put this in perspective, primary neocloud rival CoreWeave disclosed at the end of 2024 that it had more than 250,000 GPUs installed across a 360MW active power footprint. Now, CoreWeave has 470MW of active power, and 2.2GW of total power contracted with its acquisition of Core Scientific. At its current 470MW active power size, CoreWeave is nearing a $5 billion annual revenue run rate.

For a deeper look at Core Scientific and CoreWeave, read more here: CoreWeave: AI Infrastructure Built for the Next Decade; Upside Down Business Model and Core Scientific: Expects 250MW of Billable Capacity to CoreWeave by Year-End.CoreWeave: AI Infrastructure Built for the Next Decade; Upside Down Business Model and Core Scientific: Expects 250MW of Billable Capacity to CoreWeave by Year-End.

$3.75 Billion Capital Raise to Support New Growth

The day following the Microsoft deal announcement, Nebius priced a $2.75 billion convertible debt raise to support future growth, upsized from $2 billion and combined with a $1 billion share offering at $92.50. It also builds upon a recent $1 billion convertible debt raise from June, allowing Nebius to pursue rapid capacity expansion, high-quality data center site procurement and more GPU purchases.

Considering that Nebius is not as highly leveraged with debt as CoreWeave, it enjoyed significantly better terms for this raise than the other neocloud.

Nebius priced $1.375 billion in 1.0% notes due 2030, and $1.375 billion in 2.75% notes due 2032, with effective conversion prices of $159.56/share.
For comparison, CoreWeave recently closed a 9.0% $1.75 billion note in July and a 9.25% $2 billion note in May.

These attractive low coupon notes save Nebius nearly $300 million annually in debt interest payments versus CoreWeave’s recent raises, freeing up more cash for expansion or capex. However, ~$50 million in interest is nearly half of Nebius’ current quarterly revenue, adding some strain on margins though this will ease as revenue ramps.

What raises the risk for Nebius is a constant need for more capacity – it is already outpaced by CoreWeave, and simply securing >1GW of power is not nearly enough to fuel consistent growth considering some Bitcoin miners have more in their pipeline.

Capacity at scale is not cheap – for a similar 300MW data center like the New Jersey site, total development costs would likely be in the range of $8 billion to $10 billion, assuming greenfield construction costs of $9 million to $13 million per MW, and hardware costs of $18 million to $20 million per MW. Thus, adding an additional 1GW sometime in the future to take total power to 2GW (still below CoreWeave’s 2.2GW contracted), could cost an additional $20 billion.

Vertical Integration, Custom Servers, Competitive Pricing Provide an Edge

Considering CoreWeave’s broadening presence securing capacity deals with multiple Bitcoin miners, it’s important to touch upon what separates Nebius from its neocloud rival, why it may be positioned to secure future large-scale deals and extend its hypergrowth phase.

According to Nebius, the company has a few key advantages that provide it an edge when it comes to AI infrastructure and GPU rental prices:

In-house custom designed servers offer a lower TCO, up to 20% lower versus other cloud providers
Power usage effectiveness (PUE) near industry lead, leading to increased rack density per MW
Data center ownership cuts out expensive colocation fees, allowing for competitive GPU rental pricing while maintaining margins
Vertically integrated with a full-stack, proprietary AI cloud purpose-built for AI workloads

Bringing server design in-house and bypassing traditional server OEMs such as Dell and Super Micro gives Nebius control over component purchasing and rack design, helping maximize rack efficiency and minimize costs. For example, Nebius says that average power consumption for its in-house servers can be up to 20% lower versus comparable off-the-shelf servers, at 8.2 kw versus 10 kw for a fully configured 8-GPU HGX H100 server. Not only does this lower initial hardware costs to stand up data centers, but also reduces long-term power costs from lower power consumption.

Nebius also claims to offer industry-leading power-usage effectiveness of its servers, essentially on par with Azure and slightly ahead of Oracle Cloud. To understand why this is important, let’s break down PUE first: it is a ratio that compares total facility power to core IT load. For example, a 500 MW facility that can hold 400 MW of IT load (servers, cooling infra, etc.), would have a PUE of 1.25, while a 500 MW facility that can hold 450 would have a PUE of 1.11.

Lower PUE is important as it means data centers can deliver more compute per watt, with less power lost to overhead. Considering electricity is the largest operating expense for data centers, providers that can offer the lowest PUEs are more attractive as this translates into significant savings.

For example, assuming 400MW of core IT load operating in two data centers with PUE of 1.25 and 1.11, the first data center will have total power consumption of 500MW and the second 444MW. Assuming continuous operation, the lower PUE data center will consume 650,000 MWh less electricity annual, which translates to more than $90 million in annual savings at $0.14/kWh.

Source: NebiusNebius

Additionally, Nebius is prioritizing ownership of its data centers, which eliminates colocation fees, often up to 30% of total costs. Combining this with in-house designed servers and low PUE, Nebius can enjoy 20% to 25% cheaper operating costs per GPU, allowing them to offer competitive rental pricing. SemiAnalysis says that Nebius can offer “the lowest absolute price and the best terms for short to medium-term rents.”

Nebius currently offers H100 GPUs for as little as $2.00 per GPU hour with at least a three-month commitment ($2.95 per GPU hour without), a ~7% discount to the industry average at $2.15 per GPU hour, according to Silicon Data. This also compares to $2.49-$3.29 per GPU hour at Lambda and ~$6.16 per GPU hour for CoreWeave (based on $49.24 for an 8-GPU instance).

For the HGX H200, Nebius is offering rentals at $3.50 per GPU hour, a ~44% discount to CoreWeave at $6.31 per GPU, based on a $50.44 8-GPU instance price. For Blackwell GPUs, Nebius is renting HGX B200 chips at $5.50 per GPU hour, a slight premium to Lambda’s $4.99 for the B200 and a discount to CoreWeave’s $8.60 (based on a $68.80 price per 8-GPU instance).

However, it cannot be ignored that some of this pricing dynamic may stem from Nebius’ current scale, a fraction of CoreWeave’s size, and that it may cater more towards AI startups rather than larger enterprise customers who may require cluster sizes well beyond what Nebius can offer.

Nebius may also be in a tight position to demand premium pricing as bootstrapped AI startups may simply choose whatever cloud provider can offer GPU access at an affordable price, and offering competitive pricing or discounted rates can drive customer acquisition. Additionally, CoreWeave’s larger GPU fleet likely means it can demand premium pricing and long-term contracts with anchor tenants such as OpenAI, whereas Nebius is not yet at the scale to do the same beyond its Microsoft deal.

In the longer term, Nebius believes that its vertical integration with a full stack of AI services will help broaden its customer base, increase platform stickiness and capture higher margin revenue and services. Nebius offers a proprietary cloud platform with managed MLops services, low downtime and high cost efficiency, combined with its inferencing platform AI Studio. With AI Studio, Nebius says it can offer up to 3x token savings with low latency, and up to 4.5x faster time to token versus other competitors in Europe.

Nebius also reported its first MLPerf Benchmark results, training Meta’s Llama 3.1-405B model in 124.5 minutes with 1,024 Hopper GPUs, nearly double the speed of its 512 GPU result of 244.6 minutes. Nebius says that the result validated its platform’s ability to deliver bare-metal comparable performance while in the cloud.

Blackwell Ultras Drive 14% Increase in Annualized Run Rate Guidance, Not Including Microsoft Deal

In Q2, Nebius boosted its annualized run rate guidance by 14%, supported by its data center expansion, more power coming online in 2H and the rollout of more Blackwell and soon Blackwell Ultra GPUs. To note, the increased ARR guidance preceded the Microsoft deal, although it may be unlikely that ARR guidance will be raised much this year considering that capacity roll-out is more geared towards 2026.

Q2 ARR rose ~73% QoQ and 438% YoY to $439 million, marking a significant two-quarter expansion from $90 million in Q4. Nebius is targeting ARR to more than double over the next two quarters, raising its guidance to $900 million to $1.1 billion, up from its prior view for $750 million to $1 billion.

Nebius explained that the “increase in our ARR guidance reflects the strong demand we are seeing and the expected delivery of additional GPU capacity later this year, particularly the Blackwell Ultras. Because much of this capacity will come online by the end of the year, the impact will show up more in ARR than within the year’s revenue.” Much of the ARR growth is likely to be weighted towards Q4, considering the majority of GPU installations will take place that quarter, such as the ~4,000 Blackwell Ultras in the UK.

Nebius remains confident in reaching this ARR target, noting that a majority of the guidance is already under contract, while additionally capacity sells quickly upon coming online, adding an extra level of confidence.

Solidly in Multi-Year Hypergrowth Phase

Nebius is one of the more unique names in the AI universe as one of the few, if not only, AI-focused stock to have multiple years of triple-digit revenue growth ahead. Though the ARR guidance only extends two quarters ahead, it is a leading indicator of the upcoming hypergrowth phase Nebius is expected to see.

Nebius had already laid out expectations for a meaningful acceleration next year prior to the Microsoft deal, based on timing of capacity ramp in Q4 2025 and throughout 2026. The recent deal with Microsoft is extending this acceleration, with revenue revised more than $3 billion higher by 2029, or 24% to 65% above prior expectations.

Prior to the Microsoft deal, Nebius was expected to surpass $5 billion in annual revenue in 2029, doubling from $2.5 billion in 2027. After the deal, revenue estimates have been raised 24% to $3.1 billion in 2027, and 65% higher to $8.44 billion by 2029.

In terms of YoY growth, Nebius is now expected to see three consecutive years of triple digit growth from FY25 through FY27, with the possibility that additional hyperscaler deals and accelerated capacity expansion translate into a fourth consecutive year of triple-digit YoY revenue growth.

Revenue growth estimates for FY26 have moved 15 points higher to 166% YoY, while FY27 has now crossed into triple-digit territory at nearly 106% YoY. FY28 growth estimates have jumped more than 30 points to 85%, and, as noted previously, could potentially reach triple-digits if well supported by demand and capacity growth. Microsoft’s deal helps de-risk the growth story to some degree by materially upgrading backlog-like visibility to several billion in revenue through the duration of the deal.

For comparison, CoreWeave is expected to generate 128.4% YoY growth in 2026, though at a substantially larger scale with revenue estimated at $12.01 billion, before rising to $25.8 billion by 2029, or more than 3x the $8.44 billion estimate for Nebius.

Subsidiaries: Avride Progressing with Robotaxis with Key Partners Uber, Hyundai

Outside of its core AI infrastructure services, Nebius has two main subsidiaries: autonomous driving startup Avride and AI ed-tech firm TripleTen, as well as economic stakes in Toloka (now deconsolidated from results) and ClickHouse.

Avride

Avride is continuing to scale its autonomous delivery robot operations, working with Uber Eats in Jersey City, Dallas and Austin, and with Grubhub at the Ohio State University. Avride also is working with Japan’s Mitsui Fudson to deploy autonomous robots for warehouse-to-store logistics.

Avride is also advancing preparations for its autonomous ride-hailing service with Uber later this year in Dallas, based on a fleet of Hyundai Ioniq AVs with Avride handling software and systems integration. Avride said on September 4 it was beginning to ramp up testing to prepare for the Dallas rollout.

Avride is still pre-revenue and burning through cash, and due to the capital intensive nature of scaling autonomous delivery robots and AVs, Nebius is active discussions with potential strategic partners. It will be very challenging, if not impossible, to scale both AI infrastructure, which seems to be priority, with AVs simultaneously with only a few billion in cash. Consider that Waymo has raised more than $11 billion of outside capital, not including what Google has burned through.

TripleTen

TripleTen is a consumer-facing AI ed-tech platform primarily focused on offering coding or tech-skill based bootcamps. TripleTen delivered revenue of $28.8 million in 2024, up 251% YoY, while students enrolled rose 149% to 14,000.

Clickhouse

Nebius has a minority economic stake in open-source database platform ClickHouse, which recently raised $350 million in a Series C round in May. This valued the startup at $6.35 billion, more than tripling its $2 billion valuation from 2021.

Nebius has been straightforward in utilizing its 28% stake in Clickhouse as a means of raising capital, as it could unlock almost $1.8 billion (or more depending on exit valuation) without incurring dilution to shareholders or adding debt to the balance sheet.

Toloka

Toloka is a data provider for LLM and genAI developers, focusing on high-quality data solutions for training, fine-tuning, alignment, and more, counting Amazon, Anthropic, Microsoft and Shopify among its customers.

As of Q2, Nebius is deconsolidating Toloka from earnings results, as its voting share dropped below 50% following a $72 million investment from Bezos Expeditions and Shopify CTO Mikhail Parakhin. Toloka’s revenue rose 138% YoY to $26.4 million, and according to 2025’s guidance, was expected to at least double again to $50 to $70 million.

Financials

Revenue Rises 625% YoY in Q2

Before discussing the financials, it’s important to note that growth figures exclude Toloka’s impact following the deconsolidation in Q2.

Nebius reported a slight beat in Q2 with revenue of $105.1 million, up 625% YoY and 106% QoQ, versus estimates for $101.2 million. AI cloud infrastructure revenue rose more than 9x YoY in the quarter, fueled by strong demand for Hopper GPUs and almost peak GPU utilization.

Looking ahead, Q2 is marking an inflection point for revenue, with growth accelerating sequentially on a dollar basis. Q3 is estimated to see revenue up more than $52 million QoQ to $157.1 million, before rising another $106 million QoQ to $263.8 million in Q4.

For FY25, Nebius has maintained its guidance for $450 million to $630 million in revenue, excluding $50 million to $70 million in revenue attributed to Toloka. What’s interesting here is that management had explained that Nebius could grow faster, but it was “oversold on all of our supply of previous generation Hoppers, and we decided to wait for the new generation of GPUs to come. And finally, the new Blackwells are coming to the market in masses, and in parallel, we are dramatically increasing our data center capacity.” This is fueling strong growth through 2026 and beyond as these Blackwell GPUs come online.

Operating Margins Deeply Negative, Signs of Scale Emerging

While gross margin is expanding rapidly, up from 26% in Q4 to over 70% in Q2, operating margins remain deeply negative as due to the high costs of aggressively expanding data center capacity and GPUs on hand.

Excluding Toloka’s contribution, gross margin was 71.3% in Q2, up nearly 20 points from 51.5% in Q1 and improving nearly 25 points from 46.9% in the year ago quarter. While this signals strong execution to drive this degree of expansion, Nebius is now slightly below CoreWeave’s Q2 gross margin of 74.2%, and it remains to be seen how much further upside there is to gross margin as new capacity comes online.

On the infrastructure side, Nebius is pricing GPUs at a “healthy margin” on a per hour basis, while expectations for premium pricing for Blackwell GPUs may provide a tailwind through 2026. Nebius expects its Hopper GPUs to break even in two to three years on a gross profit level including both hardware and operational expenses, or even quicker when factoring in its higher-margin software and services from its full stack platform. Nebius has not commented on break-even periods for Blackwell GPUs, as it is still actively rolling out the new generation.

Moving down the line, operating margin remains deeply negative at more than (100%) of revenue as expenses continue to outpace revenue at current scale. Q2 operating margin was (105.8%), a notable improvement from (236.4%) in Q1 and (773.8%) in the year-ago quarter. Nebius is exhibiting initial signs of operating leverage, with total operating expenses up just 71% YoY versus the 625% YoY growth in revenue; this was driven mainly by 560% YoY growth in depreciation from increased server capacity, as SG&A decreased (10%) YoY.

GAAP net margin was 556%, as Nebius benefited from a one-time $597.4 million gain from revaluation of equity investments (Toloka’s deconsolidation). Adjusted net margin offers a clearer view of Nebius’ trajectory, coming in at (87.1%) in Q2, up from (164.4%) in Q1 and (424.8%) in the year ago quarter.

Quick Shift to Positive Adj EBITDA

Despite still investing heavily in capacity and its GPU fleet and kicking off its hypergrowth phase, Nebius is showing strong cost control as its core AI infrastructure business shifted to positive adjusted EBITDA in Q2. However, corporate adjusted EBITDA remains negative as Nebius continues to invest in Avride and TripleTen.

Q2 adjusted EBITDA was ($21 million), or a (20%) margin, improving from a (113.2%) margin in Q1. Nebius had laid out expectations in Q1 for adjusted EBITDA to shift to positive territory by 2H 2025, with Q3 the expected inflection point. Q2’s result does help cement this shift with higher probability now, but it will need to be proven.

For 2026, Nebius expects corporate adjusted EBITDA to be positive for the year, building on the AI cloud momentum as capacity and revenue ramp. Nebius has also offered some long-term targets for when its AI cloud business reaches scale (its medium term ‘several billion’ revenue size). At this scale, Nebius is targeting adjusted EBITDA margins between 20% to 30%, assuming a depreciation schedule of four years. Attaching this to FY29’s projected revenue of $8.44 billion, this would imply around $1.7 billion to $2.5 billion in adjusted EBITDA.

EPS Expected to Remain Negative as Nebius Scales

Nebius is expected to report negative EPS through all of fiscal 2025 and 2026, with no clear indication yet of when the company could or will shift to profitability. Current estimates show Nebius losing ($1.47) in 2025 and minimal improvement to ($1.34) in 2026.

There are no analyst estimates beyond 2026, but given the capital intensity of greenfield data center development, continuous scaling of capacity, not to mention potential investments into Avride, Nebius may face a long road to profitability.

Source: YChartsYCharts

Capex at $2B for 2025 Pressuring Cash Flows

Aggressive capacity expansion plans and a 33% increase in FY25 capex expectations from $1.5 billion to $2 billion, or nearly 4x expected revenue for the year, mean Nebius is quickly burning through cash.

Operating cash flow was ($167.7) million in Q2, for a (159.6%) margin, improving from a (357.7%) margin in Q1. However, this represented just a $30 million sequential improvement from ($197.8) million, suggesting that the path to positive cash flows may be prolonged.
Free cash flow was ($678.3) million in Q2 for a (645.4%) margin, versus ($741.8 million) for a (1341.4%) margin in Q1.
Capex was $510.6 million in Q2, or nearly 5x of revenue, down slightly from $544 million in Q1. 1H capex surpassed $1.05 billion.
Cash and equivalents totaled $1.68 billion, up only slightly from $1.45 billion in Q1 as Nebius deployed much of June’s $1 billion raise to capex and operations. Including the recent $2.75 billion raise and $1 billion share sale, cash is likely around or above $5 billion. At current burn rates, Nebius would have nearly 10 quarters of cash, but it is likely that capex will accelerate (think of CoreWeave as a leading indicator) to support growth in power and capacity.
Debt was $0.98 billion as of Q2, not including the recently priced $2.75 billion in notes.

Regarding capex, management stated that they “want to be opportunistic when it comes to really ramping up our infrastructure capacity as we see demand, and so we want to be able to sort of chase secure that demand well. And so we’ve considered some additional investments beyond the initial data center expansion plan.”

However, AI infrastructure is costly, and Nebius has already increased its guidance quite substantially for its size. The $2 billion guide also came prior to the Microsoft deal, which could add upwards force to capex and further pressure cash and cash flows, forcing Nebius into a vicious cycle of raise to build.

For example, CoreWeave is guiding to spend $20 billion to $23 billion in capex this year, also more than 4x its revenue and with potentially up to $15 billion hitting in Q4. For Nebius to keep pace with CoreWeave and scale at accelerated rates, capex needs are only going to move much higher, one that its balance sheet may not be able to sustain at the moment.

Earnings Q&A

Greenfield Site Selection

Given that Nebius is looking to acquire multiple greenfield sites for new capacity, analysts questioned why that was the route management is taking versus colocation or build-to-suit. While greenfield development provides full control over construction and design from the ground up, it can be a more expensive and longer time to power versus colocation using existing infrastructure.

Chief Product and Innovation Officer Andrey Korolenko answered why Nebius favors greenfields over build-to-suit:

“We typically favor greenfields because we can control every aspect of the data center from the design to construction to the hardware installations and deployment and phasing. We can actually tailor the phasing according to our demand. And for us, it's cheaper to build than build-to- suit, and we are not locked into the long-term leases. Also, by controlling the design of the building, starting from the — how power is piped into building and design and installation of our own racks and servers, we can achieve a lower total cost of ownership, probably around 20% less than the market average.”

Quickly Selling Through Capacity

Management also shed light on utilization trends, which support accelerated revenue growth moving forward as more capacity comes online. Higher utilization rates are important as it means GPUs are not sitting idle for long periods of time, generating revenue and shortening payback periods for high-capex server investments.

Chief Revenue Officer Marc Boroditsky explained that as Nebius “brought on more capacity, we sold through it. And by the end of the quarter, we were at peak utilization. There's a nice trend that we're actually starting to witness. As we bring on larger clusters, we are able to bring on new large customers who want to purchase greater and greater capacity. This allows us to expand and diversify our customer base and has been a clear signal there is growing opportunity in the market. This also suggests strong demand to support ramping up our capacity. If we had more capacity in the second quarter, we probably would have sold more as well.”

This ties in to comments about how growth could have been higher if Nebius was not GPU constrained, and signals a healthy demand environment for the moment as inference begins to take off. However, this is not a trend that can be assumed to last forever, given the competitive nature of the neocloud and hypercloud arena and the fact that Nebius’ growth is directly tied to its ability to deliver increasingly more capacity without delay.

Tariff Impacts:

Interestingly, Nebius discussed potential tariff impacts on Q2’s call, noting that it is still a bit early to form any definitive conclusions regarding any detrimental impacts. Chief Communications Officer Tom Blackwell said that “for now, it's a bit early to say anything definitive” about how tariffs could affect growth, though he stated that it is “possible we could potentially see some short-term fluctuations.”

Conclusion

Neoclouds present a high risk/high reward opportunity for investors, as mega-deals like Microsoft’s recent contract with Nebius highlight just how supply constrained the market is. On the other hand these firms have little access to organic cash and cash flows, and instead must turn to debt to fund aggressive capacity expansion.

The Microsoft deal and aggressive capacity expansion are fueling a prolonged period of hypergrowth for Nebius, with the company currently expected to see three, potentially four, consecutive years of triple-digit revenue growth. However, capex is outpacing revenue by a factor of 5x, and the recent debt and share sale may only provide a few quarters of runway before more cash is needed to fund this growth.

As stated in the intro, ultimately, we are stepping away from the stock. While it’s clear that Nebius could do well in an AI-driven market and the stock could extend rapidly, we want to be prudent about what will hold up should we see a softer AI narrative. We like other names in our portfolio better in terms of participating in the upside while protecting to the downside for when we do get that inevitable tech selloff. Primarily, the negative free cash flow, capital intensive operations and weak margins are among reasons we believe there are stronger names in our portfolio.

Please note: The I/O Fund conducts research and draws conclusions for the company’s portfolio. We then share that information with our readers and offer real-time trade notifications. This is not a guarantee of a stock’s performance and it is not financial advice. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.

Recommended Reading:

Supermicro Fails to impress; Volatility is here to staySupermicro Fails to impress; Volatility is here to stay
Bloom Energy: First Direct Hyperscaler with Oracle; More are Likely to FollowBloom Energy: First Direct Hyperscaler with Oracle; More are Likely to Follow
Positions Report – July 2025Positions Report – July 2025
AMD Reports in Line while AI Story to Improve from HereAMD Reports in Line while AI Story to Improve from Here

Aehr Test Systems: Optimism Driven by Sonoma Follow-on Orders Despite Soft FY25

Posted on September 12, 2025June 30, 2026 by io-fund

Aehr is a familiar stock to those who have been with the I/O Fund for some time. We’ve held the stock in the past, participating in the company’s former upward trajectory from wafer testing and burn-in systems that stress test devices to avoid early failures.

While Aehr has struggled as of late, plagued by weakness in EVs weighing heavily on SiC revenue, the company is working to pivot for wafer testing and burn-in of AI processors following last year’s acquisition of Incal. Though Aehr’s stock has witnessed a sharp rally since the summer following a handful of AI orders, growth in FY26 remains pressured by SiC weakness as AI begins to ramp up.

Background on AEHR:

The primary market where AEHR saw former success was with electric vehicles (EVs) as Tesla, for example, made the switch from Si-IGBT (silicon-insulated bipolar transistors) to silicon carbide MOSFETs for EV components. By switching to silicon carbide (SiC), Tesla was able to build traction inverters, DC/DC inverters, on-board chargers, fast chargers and energy storage applications that charged faster and offered a longer range of miles. We’ve covered this in the past here.

As you can imagine, Aehr's stock has struggled over the past two years as EV sales have declined, causing companies like ON Semi (who supplies Tesla) to cut their orders with Aehr.

Yet, Aehr’s stock has seen a resurgence as of late based on its burn-in solutions for AI processors. Similar to EVs, by stress testing the chips at elevated temperatures and voltages (“burn in”) at the wafer-level, Aehr can lower costs from failures happening at the package or system level. Overall, Aehr’s value proposition is to make sure expensive chips don’t fail when placed under stress.

Although silicon carbide is becoming a lower percentage of Aehr’s overall revenue as they shift toward AI processors, it’s Aehr’s background with silicon carbide that is becoming useful for AI data centers to perform testing to prevent overheating or other failures at the wafer level. Per the CEO: “Many AI processor companies are talking about billions of dollars of devices a year with the largest AI processor company in the world shipping over $100 billion worth of processors in the data center applications this year alone. Even a 0.1% increase in yield by shifting the burn-in of devices from the system or heterogeneous package level to wafer level is very significant.”

Testing at the wafer level is attractive due to the complexities of AI hardware, as advanced packages combine multiple GPU or ASIC dies (including dual-die setups) with 8- to 12-high HBM stacks—often totaling dozens of memory dies, all assembled via technologies such as CoWoS. This complexity increases as AI processors reach the reticle limit (maximum area that can be exposed in a single path of lithography equipment). By using chiplets, or smaller dies, to form a larger system, the reticle limit is circumvented for a larger transistor count. This places more emphasis on CoWoS technologies to integrate multiple chiplets onto an interposer.

When Nvidia attempted to use CoWoS-L packaging, there were reported delays due to “alleged mismatch in the coefficient of thermal expansion (CTE) among the GPU chiplets, LSI bridges, RDL interposer, and motherboard substrate led to warping and system failure.”

What that describes is that AI processors and advanced packaging could benefit from earlier testing as these delays of about 3 months in production have led to nearly 6-9 months in delayed timelines for shipping in volume.

InCal Acquisition and Sonoma Systems

Aehr announced its acquisition of Incal just over a year ago, buying the AI semiconductor burn-in test solutions manufacturer for total consideration of $21 million, or ~1.75x Incal’s TTM revenue of ~$12 million at the time. While the deal was financially accretive on the top-line, more importantly, it is Incal’s Sonoma Test System that is the primary product driving Aehr’s AI and HPC transition. The high-powered system is used for test and burn-in of leading-edge AI processors/GPUs and networking chips up to 2,000 watts.

Combined with Aehr’s FOX systems, Aehr is now the only company that can offer wafer-level and package part burn-in for qualification and production of AI processers. Additionally, Aehr can offer prospective customers direct, side-by-side comparisons for testing costs, output, operational costs, and impact on yields, translating to improved results at the manufacturing level and higher revenue and profits from improved yield.

At the time of the acquisition, Aehr stated that it believed that its manufacturing capacity and R&D resources would help accelerate production and adoption of Incal’s Sonoma Systems in the AI market. Aehr backed this up in Q4 by stating that it has shipped more Sonoma systems post-closing than Incal had in the prior three years.

In Q4, Aehr disclosed that it has upgraded its facility to be able to manufacture ten to 20 systems at once, if needed, should demand require them to produce multiple systems at once for different customer shipments. Aehr said that it won its first production AI processor customer during the fiscal year and received “initial volume production orders for the multiple Sonoma ultra-high-power system.” This customer, while unnamed, was stated to be a premier data center hyperscaler producing their own AI processors and ramping capacity significantly.

More orders for the Sonoma systems have been a core factor in Aehr’s recent multi-month rally, as the broader opportunity for wafer-level and package part burn-in for AI is multiples larger than SiC. These orders are instilling a higher degree of confidence in Aehr’s ability to transition to a higher AI mix while navigating the difficulties of SiC.

Silicon Photonics ICs: Another Upcoming Market

Another emerging opportunity for Aehr is the silicon photonic IC market, as adoption of optical chip-to-chip and optical networking switches rises with Nvidia, AMD, Intel, TSMC and GlobalFoundries all announcing roadmaps featuring optical chip-to-chip communication.

Silicon photonics are expected to be the only viable choice for rack-to-rack interconnects and across the data center due to the need for high bandwidth and lower power at high speeds. There is also low-loss over long distances with optical fiber, which refers to preserving the original signal, whereas copper sees signal degradation over longer distances. For example, Nvidia stated that by replacing pluggable optics with silicon photonics on the package, it can “deliver 3.5x more power efficiency, 63x greater signal integrity, 10x better network resiliency at scale and 1.3x faster deployment compared with traditional methods.”

Aehr says it has five to six customers in the SiPho space, some noted above, with one of the customers being an OSAT (outsourced semiconductor assembly and test) that purchases Aehr’s systems. Management said that they have seen a “significant number of new wafer pack designs from our installed base of systems…that they use for qualification and development work on their FOX wafer level test and burn-in systems.”

Additionally, Aehr stated that it now is offering a new higher-power system, up to 3,500 watts per wafer, to meet the higher power needs for optical I/O and chip-to-chip communication devices. The new system is available as an upgrade to FOX-NP systems for low-volume production and to the FOX-XP 9-wafer system for higher-volume production.

However, while SiPho is emerging as an entirely new market to capture, the long-term opportunity remains rather limited, with management forecasting SiPho’s opportunity to be smaller than SiC by the end of the decade.

Recent AI Orders Rejuvenate Shares with 63% Rally

On July 22nd, Aehr shared a press release in which it announced orders for 8 Sonoma systems from its lead production AI-processor hyperscale customer, with delivery expected to occur over the next 2-3 quarters. On August 26th, an additional press releaseadditional press release announced a follow-on order for 6 more Sonoma systems from the same hyperscaler, with delivery scheduled across the next 2 quarters (likely contributing to FY26 H1).

This concurred with a paid evaluation of Aehr’s Fox-XP systems from a leading AI processor supplier, announced on August 25. Aehr said this 3-6 month paid evaluation features a custom WaferPak high-power wafer contactor and a production wafer-level burn-in test program development. Management noted that while they cannot guarantee a final purchase, they believe this marks the first step toward adoption of its WLBI solutions as an alternative to this customer’s production burn-in done in later manufacturing. They also believe that a successful evaluation phase could quickly transfer into high-volume production, which they see as a significant growth opportunity. Notably, Aehr has already proven to some degree that its FOX-XP system is viable for high-volume test and burn-in for AI processors, having received a >$10 million order in December 2024 from a leading AI accelerator company.

Tracking these AI orders and related revenue is important as AI is now a much larger portion of Aehr’s revenue, both from the acquisition of Incal and the SiC slowdown. For example, AI accounted for 0% of Aehr’s revenue in FY24, but more than 35% in FY25, and these orders suggest AI’s contribution could increase further in FY26 as deliveries occur.

Despite weak financial results reported in FY25, 14 Sonoma systems ordered over a 2-month span by a single customer helps validate the efficacy of the tech. While underwhelmed with the recently reported results discussed further below, investors have now turned optimistic due to these repeat orders & the potential for revenue growth.

2026 Revenue Estimates Getting Crushed

This flurry of orders has rejuvenated Aehr’s stock with shares rising 63% since the first order announcement in July, signaling increased optimism in Aehr’s ability to capture more AI-related growth as the silicon carbide market lags. This compares to less than a 2% gain for the Nasdaq 100 over the same period.

The big question here for Aehr and for investors likely hinges on one key facet: can these new AI orders help drive a meaningful inflection in revenue to make this run sustainable. Since April, Aehr’s valuation has tripled, and shares no longer appear cheap considering FY26 revenue estimates have gotten crushed even after a weak FY25.

For FY26, revenue estimates have plunged from $92 million in Oct 2024 to $73 million in April 2025 and now barely $61 million in August. This would correspond to minimal 4% YoY growth after a challenging FY25 where revenue declined nearly (11%) YoY. It’s important to note that visibility on Aehr’s future growth can be very limited with minimal analyst coverage, so estimates can change quickly, such as if Aehr secures a large order.

What would be crucial to see here is if these recent AI orders can help drive revenue for the year higher, or if SiC headwinds will weigh on growth. SiC had accounted for >90% of revenue in FY24, though it declined sharply to <40% in FY25, or a decline of more than (56%) YoY. Keep in mind that Aehr also warned that SiC may not experience order growth in fiscal 2026 as “customer forecasts for this market are back-half loaded, with stronger growth expected in our fiscal 2027.”

Timing Issues Impact Q4 Revenue, Yet Rebound is Prolonged

Fiscal 2025 revenue declined (10.9%) YoY to $59.0 million, versus FY24 revenue of $66.2 million, reflecting a continued SiC slowdown and tougher prior-year comps, with Q4’24 being a very strong quarter due to elevated EV chip demand. Revenue declines in Systems (-9% YoY) and Contractors (-18% YoY) were partially offset by increase in Services +37% YoY as Incal acquisition contributed $18.6M to FY25 post close, up more than 50% YoY. While SiC / EV revenue weakened, early signs of diversifications towards AI, HDD, services, and US mix helped dampen the decline.

Q4’25 was also tough with revenue of $14.1 million, a (23%) decline against Q3’25 revenue of $18.3M. Year over year figures reflect the same softness, a 15.1% decline of $2.5M YoY from Q4’24 revenue of $16.6M. Management noted the YoY decrease was “primarily due to a delayed shipment of a FOX-CP system that was forecasted to be shipped to our hard disk drive customer. Because of tariff-related uncertainties, probers sourced from Asia to support the FOX-CP system were delayed. We now expect to complete this shipment in our current quarter, Q1 of fiscal 2026.”

Beyond this timing issue, product-line trends provide additional insight into the revenue composition. WaferPak revenues were $4.2 million and accounted for 30% of Q4 revenue, underscoring Aehr’s ongoing pivot toward packaged-part burn-in. Package part burn-in now represents nearly half of quarterly revenue, partially offsetting legacy wafer-level volatility.: “Sonoma, Tahoe and ACO package part burn-in systems continue to contribute strongly, accounting for 44% of our fourth quarter revenue. .” The takeaway here Is that Q4's revenue contraction reflects end-market and timing headwinds, not necessarily competitive erosion – positioning Aehr for potential rebound as deferred shipments clear and packaged-part momentum continues.

As we look out to FY26, analysts see Q1 revenue of $11.46 million, a decline of both (20%) QoQ and (16%) YoY. For Q2, analysts see revenue of $14.4 million, signaling a pick up to 26% QoQ but just 7% YoY. Coming into the year, analysts originally expected 30% YoY growth for FY26. That number has now fizzled down to a measly 4%. Management would argue that a strong pipeline in AI and memory could make FY26 a different story.

Geographically, Aehr was able to generate YoY revenue growth in Q4’25 from the US ($4.1 million vs $3.8 million) and Europe ($1.3 million versus $0.9 million). Unfortunately, the sharp decline in Asia revenue ($8.7 million vs $12.9 million) significantly offset any of these incremental improvements. Management noted that the Asia softness was driven by tariff-related uncertainties (e.g. sourcing disruption in Asia) causing delayed shipment of FOX-CP System, although shipment is now expected to occur in Q1’26.

Key Segments

Contractors Revenue:

Q4’25 vs Q3’25 (QoQ): $7.35 million vs $9.91 million (Q3’25) represents a (25.8%) decline
Q4 FY25 vs Q4 FY24 (YoY): $7.35 million vs $9.80 million represents a (25%) decline
FY25 vs FY24 (YoY): $30.8 million vs $37.5 million (FY24) represents a (17.9%) decline

Contractors revenue was the most cyclical and the biggest drag on QoQ and YoY performance. Contractors scale with system utilization – less wafers going through Aehr equaling lower contractor demand.

Systems Revenue:

Q4’25 vs Q3’25 (QoQ): $4.78 million vs $6.28 million (Q3’25) represents a (23.9%) decline
Q4 FY25 vs Q4 FY24 (YoY): $4.78 million vs $4.96 million (Q4’24) represents a (3.6%) decline
FY25 vs FY24 (YoY): $22.0 million vs $24.2 million (FY24) represents a (9.1%) decline.

Q4 systems shipments were lighter than Q3 which reflects timing of customer evaluations and slower follow-on orders from SiC. System revenue will be lumpy and fluctuate quarter to quarter, with the full year decline tied to fewer installs in the Asia/ EV end market.

Services Revenue

Q4’25 vs Q3’25 (QoQ): $1.96 million vs $2.08 million (Q3’25) represents a (5.8%) decline.
Q4 FY25 vs Q4 FY24 (YoY): $1.96 million vs $1.84 million (Q4’24) represents a 6.5% increase.
FY25 vs FY24 (YoY): $6.14 million vs $4.49 million (FY24) represents a 36.8% increase.

Service revenues are contract-driven and therefore recurring and less volatile in nature. While down QoQ, the slight growth in YoY metrics noted above shows stickiness with existing customer base and early AI evaluation work. While systems and contractor revenue swings, look for services revenue to act as a stabilizer.

Margins Show Sharp YoY Contraction

Q4’25 Adjusted Gross Profit came in at $4.89 million, reflecting an adjusted GM of 34.7%, down 42.7% reported in Q3’25. YoY figures reflect even more weakness, down significantly from 51.5% reported in Q4’24. Q4’25 gross margin marks the 3rd sequential quarter of margin degradation.

Management noted margin softness in the quarter was partially driven “high manufacturing overhead due to under absorption as our manufacturing capacity utilization was lower during the renovation of our Fremont site and the consolidation of inventory from the Incal facility.” In simpler terms, certain factory costs can either be (i) expensed right away as period costs in the P&L, or (ii) be capitalized into inventory and only hit the P&L when that inventory is sold. When production runs at normal levels, more of those costs are absorbed into inventory. But when volumes fall (like during down-time or site renovations), fewer costs are absorbed into product, meaning a bigger share flows straight to expense in the current quarter – reducing margins.

FY2025 Adjusted Gross Profit was $23.9 million versus FY2025 Gross Profit of $32.5 million, down (26.4%), reflecting a margin of ~40.6% (down ~8.5% YoY). Compared to FY2024, the margin compression & overall decline in gross profit can be attributed to: (1) volume pressure (2) inventory step-up amortization related to Incal acquisition and (3) mix shift toward lower-margin packaged part-systems. Until the backlog can convert from orders to shipments and contribute to acceleration in top line growth, expect a continued trend of lack-luster margin performance.

Q4’25 Adjusted Operating loss was ($0.55 million), reflecting a –3.9% operating margin. The QoQ decline from 8.2% GM in Q3’25 is due to the gross margin squeeze mentioned above in combination with ~$0.86M in restructuring charges. The YoY decline from 20.60% is attributed to a handful of factors: prior-year was profitable, the current year includes Incal integration, lower gross profit, and higher R&D pend for AI systems.

FY25 Adjusted Operating Loss of ($4.3M) declined substantially YoY, down $15.8M from $10.1M reported in FY24. This was also driven by the lower gross profit, higher R&D / SG&A spend tied to AI systems and Incal acquisition integration costs mentioned above. Operating leverage turned negative as opex remained relatively flat while revenue declined.

EPS Back to Negative in Q4

Q4’25 Net Income of ($0.25 million), down from $1.9 million in Q3’25 and $24.7 million in Q4’24. Q4 net margin came in at (1.8%), down significantly from the 10.8% reported in Q3. The swing downward should not be considered a seasonal dip as it reflects timing issues and margin compression. YoY comps are skewed by the $20M one-time tax benefit but reflect the same weakness. In future quarters, watch for these margins to recover with volumes – if they don’t that signals additional issues beyond timing noise.

Balance Sheet & Cash Flow

Cash & Equivalents: $24.5M, Down $4.9M or ~16.6% QoQ; Down -$24.6M or –50% YoY. On a quarterly basis, the decline is driven by cash usage from negative OCF (-$7.4M) and higher capitalized expenditures. Q4 inventory build flattened while receivables and order timing consumed cash. On an annual basis, this decline is driven mainly by cumulative FCF burn in FY25 of -$12.4M along with Incal acquisition related spend. The takeaway here is that the Company has plenty of runway with ~$25M in cash and little debt. A re-acceleration in revenue could help the company avoid having to tap into its $100M shelf.

Inventory: $42.0M, Modest QoQ decrease (–0.8%), up ~12% YoY. Inventory build continued for anticipated new orders (AI, HDD, NAND) even while SiC slowed. We will continue to monitor these levels to understand how the new systems are ramping. If SiC continues to slow which could cause inventory to become excess / obsolete.

Bookings for Q4 were $11.1M, less than half of Q3’s $24.1M. Backlog slipped to Q4 were $15.2M compared to $18.2M in PQ. With $14.1M in revenues for Q4, bookings did not fully cover sales, requiring backlog drawdown to support revenue. A disappointing performance compared to Q3 where bookings of $24.1M comfortably exceeded revenue of $18.3M. Management noted “while there was only a small amount of revenue in the fiscal year from wafer level burn-in in Hard Disk Drive components, about 10% of our order bookings for FY25 came from this new market, all of which we expect to ship and generate revenue from during this fiscal year now, '26.”

Effective Backlog (to include bookings received after quarter-end) of $16.3 million.

Net Working Capital: $73.1 million (–3.3% QoQ; –16.3% YoY). The decline in net working capital is linked to AR & AP, as order timing and delayed FOX-CP shipment affect both cash collection and payables alignment.

Debt: None, aside from lease obligations (ST $0.91 million; LT $9.92 million).

Operating Cash Flow: -$7.4M (vs. +$1.2M prior); margin: –16.3%. This is largely driven by weak earnings performance but also compounded by the consumption of working capital. Cash burn levels remain manageable when compared to liquidity, signaling that AEHR doesn’t appear to be in near-term distress.

CapEx: $5.0M (vs. $0.75M) is elevated due to Incal integration and capacity expansion.

Free Cash Flow: –$12.4M (vs. +$1.0M); FCF margin: –21.0%. These stats mentioned above reflect poor financial performance in a transitional year that included high integration costs. An improvement in future cash generation could be driven by (1) shipment of the delayed FOX-CP or (2) a ramp in AI/HDD/NAND.

Earnings Q&A:

AI Market is 3-5X Larger than SiC/EV Market for Aehr

The flurry of orders and resurgence of optimism tied to a handful of Sonoma orders is underpinned by the fact that the AI market is 3x to 5x larger than SiC, where strong growth in 2021 drove a >10x increase in Aehr’s stock within the year. Thus, the relative size of the AI opportunity theoretically could open the door to more explosive growth in the future as capex on test and burn equipment is attached to a much larger device base with strong forward growth prospects.

Aehr laid out a tentative discussion on the TAM that they believe they have in AI processors versus SiC:

“The original silicon carbide models that took a look at, say, the target applications for silicon carbide, which were primarily the electric vehicles, how many EVs, how many components would be in it, et cetera, et cetera, you could come up with how many wafer starts that would require in, say, 2030. And I know that you had put some models together at that time. There were about 4 million wafer starts. We looked at 12-hour burn-in time, single insertion with our systems. Long story short, we saw that the total market was somewhere 300, 350 of our systems with ASPs about $4 million a piece or something like that.”

Back of the napkin math here places SiC’s 2030 modeled TAM at $1.2 billion to $1.4 billion. For AI, management said that by 2030, wafer starts may actually end up around half that of SiC, but because these are 300mm wafers, up to 20,000 watts of power, they require multiple touchdowns as testing is only done at 3,000 to 4,000 watts at a time. This is where management sees the market at 3x to 5x SiC, or a $3.6 billion to $7 billion TAM.

Potential For More Incoming Orders:

As we have seen with the recent orders in August, the potential for more intra-quarter orders is a key factor in moving Aehr’s stock. Aehr does have the paid evaluation that they expect to complete over the next one to two quarters, with the possibility of transitioning to high-volume production afterwards; management hinted in Q4’s call that the decision could be made within six months with orders thereafter. They did not quantify potential sizing, but expect it to be a “significant opportunity” should it reach high-volume production as this customer’s capacity requirements are significant.

Outside of this, management stated in the call that they do expect more evaluation phases with other AI companies this year, allowing them to capture a “meaningful share” of the AI processor burn-in market with its FOX systems and WaferPak contactors.

Analysts also asked about Aehr’s first AI customer, understood to be behind the December 2024 $10 million order:

Larry Edward Chlebina

Gayn, that first AI customer at the OSAT, so they're — are you under the belief that they're really pleased with it? And do you expect more orders from them in the near future?

Gayn Erickson

Yes to both of those. Wait, you said near future. I want to be careful of setting any time lines, but I'll go out and say we expect more — just more this year, though.

Larry Edward Chlebina

And then now you have another AI customer in evaluation. So that's the second one for wafer level burn-in. And then you have a third one that's going after the production in the package part burn-in. Are they 3 distinct AI customers? Or are they…

Gayn Erickson

Yes, totally different.

Update on Customers & Concentration Easing

Aehr also provided some insight into current and prospective customers and customer concentration, noting that expansion into new markets is leading to customer concentration easing.

Christian David Schwab

Gayn, I've received a lot of questions regarding your most recent slide in your investor deck with a lot of well-known marquee names. And I'm just — is that — should investors think of that list as a list of current and previous customers? Or does it include maybe names of prospective customers such as new AI customers that you're working with or new silicon photonics customers, et cetera? How should we be thinking about that slide?

Gayn Erickson

So we're — yes, we're — the new SEC rules do not require you to name it. So unless we already have prior arranged agreements with the customers to name them, we're no longer doing that. Prior to that SEC rule, we could name them even if the customers objected, if you will, but we're not doing that now.

Below is Aehr’s slide naming its global customers, which it says is a partial list – some of the top names include Nvidia, Google, Microsoft, Marvell & Inphi, Samsung, TSMC, Broadcom and Qualcomm.

For customer concentration, Aehr said that it now has three companies representing >10% of revenue for FY25, with two of these customers “representing new markets and customers.

Tariff Uncertainty Still Lingering

Aehr has been fairly open about the effects of recent fluctuating tariff policy on results, providing commentary on tariff-impacted order timing. Aehr also temporarily withdrew its guidance as of Q3 and has not yet reinstated the figure.

In Q4’s call, Aehr said that in April they were primarily concerned with the “potential secondary impacts on our current and prospective customers as well as the possibility of pauses or delays in customer orders, shipments or supply chain deliveries.” Aehr also said that they “know it must be a broken record to hear terms like uncertainty around tariffs on many company earnings calls, but this is still the case.”

Management followed this up by saying that they still seeing tariff-related impacts on specific order timing, particularly for fiscal Q1. Additionally, there was more clarity on the delayed FOX-CP shipment to its Japanese HDD customer. Aehr had expected to receive a high-power prober part shipment for its FOX-CP systems by the end of May, yet the first shipment was not received until early July, delaying delivery.

Conclusion

Aehr’s AI pivot has renewed optimism in the stock, with AI now contributing 35% of revenue for FY25 with more orders suggesting growth in AI can continue. However, Aehr’s primary market in SiC remains weak, and management has noted that SiC may not grow in FY26, with revenue estimates plunging over the past few months as a result.

Overall, the AI processor market opens the door to a much larger TAM for Aehr’s systems compared to SiC, with management believing it to be 3x to 5x larger by 2030. Should Aehr successfully pivot to AI processor test and burn-in, and continue to scale capacity, production and orders, it can pave the way for stronger revenue growth in FY27 and beyond.

Every Thursday at 4:30 pm Eastern, the I/O Fund team holds a webinar for premium members to discuss how to navigate the broad market, as well as various stock and crypto entries and exits. Beth Kindig offers weekly deep dives including lesser-known cryptocurrencies and AI stocks, plus the team offers trade alerts. The I/O Fund team is one of the only audited portfolios available to individual investors. To receive $100 off our Advanced tier use ADVANCED100 or click here and email your request to upgrade.4:30 pm Eastern, the I/O Fund team holds a webinar for premium members to discuss how to navigate the broad market, as well as various stock and crypto entries and exits. Beth Kindig offers weekly deep dives including lesser-known cryptocurrencies and AI stocks, plus the team offers trade alerts. The I/O Fund team is one of the only audited portfolios available to individual investors. To receive $100 off our Advanced tier use ADVANCED100 or click hereclick here and email your request to upgrade.

Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.

Recommended Reading:

Oracle Soars After Earnings – Is ORCL Stock Still a Buy?

Posted on September 11, 2025June 30, 2026 by io-fund

This quarter was less about the headline P&L figures and more about a narrative shift: Oracle’s stockis being re-rated by the market as a hyperscaler and AI infrastructure play, not a slow-growth enterprise software company. The 36% gap reflects a structural repricing based on unprecedented backlog visibility and OCI’s growth trajectory.

The market is clearly excited about this report, and for good reason. Remaining performance obligations (RPO) grew 359% YoY with cloud RPO growing “nearly 500%” on top of 83% growth last year. This compares to RPO growth of 41% YoY last quarter and cloud RPO growth of 83% last year. The RPO growth was strong enough to negate the miss Oracle reported in the current quarter.

Another key reason that Oracle’s stock is exploding higher despite a lagging fiscal Q1, is that Oracle Cloud Infrastructure (OCI) was forecast to “grow 77% to $18 billion this fiscal year and then increase to $32 billion, $73 billion, $114 billion and $144 billion over the following 4 years.” You can think of this as an acceleration from roughly 50% growth on IaaS in recent quarters to up to 128% growth in future years, specifically from the $32B to $73B in the medium-term of two years out. Also, consider that Azure is at $75 billion, which means Oracle and Microsoft will likely be on par with each other over a four-year period for their infrastructure segments.

Given this, there’s no denying that Oracle reported impressive results, yet can the stock continue its run – or has the easy money been made? Below, we look more closely at Oracle’s report plus offer a buy plan that discusses how the I/O Fund – a leading AI tech portfolio – is approaching this stock.

RPO Surges to 359% YoY Growth, up from 83% Last Quarter

Remaining Performance Obligations (RPO) exploded to $455 billion, up 359% YoY. Management also guided that it could exceed $500 billion in the near term with additional multi-billion-dollar contracts, stating: “Over the next few months, We expect to sign-up several additional multi-billion-dollar customers and RPO is likely to exceed half-a-trillion dollars.” This backlog provides unprecedented revenue visibility, locking in future OCI and SaaS growth.

Oracle reported surging RPO, causing the stock to go up 36% from the earnings report.

Oracle Cloud Infrastructure (OCI) Could Surpass the Big 3 with 77% Growth

OCI (IaaS) revenue grew 55% YoY to $3.3 billion, faster than hyperscaler peers. We had pointed out in the analysis “Can Oracle Become the Next $1 Trillion AI Stock” that Oracle was quietly sneaking up on AWS, Azure and Google Cloud with a higher growth rate, stating:

“Though Oracle is growing off a much smaller cloud base than say Azure, robust IaaS momentum could drive its Cloud growth at a much faster rate than the Big 3 – defined as Microsoft, Amazon, and Alphabet — over the next few years.

As stated above, consensus currently models in $46 billion in IaaS revenue in FY28. For the IaaS segment to increase 4.5x from FY25’s $10.2 billion in revenue, this requires growth at a 65.2% CAGR, or a slight deceleration from >70% YoY in FY26 to >60% YoY in both FY27 and FY28.

This rapid IaaS growth could fuel a 40% CAGR for Oracle’s total Cloud growth by FY28, taking its Cloud segment from $24.4 billion to $66 billion. This 40% CAGR will far outpace AWS’ growth in the high-teens, and Google Cloud and Azure in the high-20% to low-30% range.”

The earnings report is especially groundbreaking as Oracle is now forecasting an even higher number than our previous models had indicated. Management raised its guidance to +77% growth in FY26 to $18B, with a clear multi-year ramp to $144 billion within five years.

Our firm had begun to form a picture that Oracle could become more attractive than the Big 3 based on previous estimates of $46 billion in FY2028. As of last night, these estimates are now at $73 billion – suggesting Oracle is becoming a force to contend with.

Oracle’s infrastructure-as-a-service segment (IaaS) is growing faster than the Big 3, putting Oracle’s revenue on a path to contend with traditional hyperscalers.

MultiCloud DB up 1,529%YoY

Over the past 1-2 years, Oracle has been working with AWS, Azure and Google Cloud to offer database collocation, running Exadata and Oracle Autonomous Database within their cloud infrastructure. This allows enterprises to leverage Oracle’s database platforms across a multi-cloud environment without major migrations and lower data egress costs This led to multicloud database revenue with Amazon, Google, and Microsoft surging 1,529% YoY.

Oracle discussed ways that multicloud will continue to grow, citing that multicloud will continue to grow from 37 data centers to a total of 71 over the next several years. These data centers host Oracle’s databases that are embedded into the Big 3 with low latency.

Oracle’s Fiscal Q1 Earnings Results

Oracle delivered Q1 revenue of $14.9 billion, growing 12% YoY but slipping 6% sequentially, coming in just shy of the Street’s $15.0 billion estimate. While the headline miss and QoQ contraction might have raised eyebrows in another context, management quickly shifted the focus forward, guiding Q2 revenue growth of 12-14% YoY – an outlook that suggests a rebound and signals confidence that momentum will accelerate from Q1 levels.

Segment Level Results: All Eyes on Cloud

Diving deeper, Oracle’s segment level results highlight a company amid a decisive mix shift. Cloud is the clear growth engine, with revenue climbing to $7.2 billion, up a robust 28% YoY and 7.3% sequentially from $6.7 billion in Q4. Cloud now represents 48% of total revenue, a sharp step up from 42% a year ago – underscoring the Company accelerating mix shift toward next-gen infrastructure and applications. Within the segment, Cloud Infrastructure (IaaS) stood out with $3.4 billion in revenue (+12% QoQ), sustaining hyperscaler-like momentum, while Cloud Applications (SaaS) delivered a steady $3.8 billion (+4% QoQ), anchored by Fusion ERP and NetSuite growth. Together, these results reinforce that Oracle’s transformation is no longer aspiration – the company is increasingly defined by Cloud, not legacy software.

Oracle’s Software revenue came in at $5.7 billion in Q1’26, essentially flat YoY but sharply lower sequentially, making it the primary drag on total revenue this quarter. The decline underscores the continued erosion in legacy licensing which remains a structural headwind. While maintenance and support revenues provide a degree of stability, the segment is steadily losing relevance in the growth narrative. Investors should expect Software to remain a transition burden until Oracle’s cloud scale-up fully eclipses the legacy base.

Oracle’s Hardware revenue was $670 million in Q1’26, essentially flat YoY (+2%) but down 21% sequentially from Q4. While the sharp QoQ decline is a visible drag on top-line optics, Hardware now represents less than 5% of total revenue and continues to shrink in strategic importance. Management has long signaled that the mix shift away from on-prem hardware is deliberate, freeing up resources to scale higher-growth cloud infrastructure. For investors, Hardware is best viewed as a legacy headwind that will gradually fade from relevance, with little bearing on the core thesis around OCI and Saas growth.

Oracle’s Services Revenue reached $1.35 billion in Q1’26, up 7% YoY from $1.27 billion but essentially flat sequentially. This segment provides steady, recurring revenue, anchored by consulting, support, and implementation work tied to Oracle’s enterprise base. While not a growth engine, Services play an important supporting role in the broader cloud story, helping customers migrate workloads and deepen adoption of Oracle’s SaaS and OCI platforms. The real value here is its stickiness – ensuring that once customers enter Oracle’s ecosystem, they are more likely to expand and consume additional cloud services over time.

EPS Split: GAAP Miss, Adjusted Beat

EPS Trends reflect a split narrative, as GAAP EPS missed while non-GAAP EPS beat. Both GAAP and adjusted EPS fell QoQ as the Company leaned heavily into investment mode. Adjusted EPS still showed positive YoY growth, flexing underlying profitability despite the cloud buildout. The market appears to have looked past the GAAP miss because adjusted EPS and cloud momentum underscore the long-term growth story.

GAAP EPS of $1.01, down 15% QoQ from $1.19 in Q4’25 and flat YoY vs. $1.03 in Q1’FY25. This figure was also lower than the analyst estimates of $1.04. This decline was largely driven by restructuring charges, higher interest expense, and heavy investment.
Non-GAAP EPS of $1.37, up 6% YoY from $1.39 in Q1’25 but down 14% QoQ from $1.70 in Q4’25.

Liquidity Steady, Capex Heavy as Oracle funds the Cloud Buildout

Oracle’s balance sheet and cash flow metrics show a deliberate tilt toward aggressive investment as the company is pulling every lever (e.g. heavy capex, payables management, heavy debt burden, etc.) in the near term to build datacenter capacity and capture OCI demand. Operating Cash flow is growing and outpacing revenue growth slightly, signaling operational effectiveness and economies of scale. Liquidity remains adequate with cash steady at $10B and deferred revenues providing visibility, but working capital is increasingly cloud-contract driven. The tradeoff here is clear: short term FCF pain for long-term hyperscaler positioning.

Next week, we will break down how capex compares to AI revenue for the major hyperscalers plus Oracle – which stock is seeing the highest ROI? Sign up here to get this free analysis in your inbox.Sign up here to get this free analysis in your inbox.

How Oracle Compares to the Big 3

Oracle’s ability to drive lower latency and high performance is one of the main reasons enterprises use Oracle for AI, as it allows enterprise customers to run demanding AI workloads faster and at a lower cost.

RDMA (Remote Direct Memory Access) is helping to drive Oracle’s AI story by enabling direct memory access between servers without utilizing CPUs, resulting in low-latency, high-bandwidth performance. Bypassing the CPU greatly accelerates data transfer rates, a necessity for large AI workloads requiring massive compute.

RDMA is integral to Oracle Cloud Infrastructure as the backbone of Oracle’s Gen2 Cloud and increasingly large Superclusters for AI training and inference, allowing ultrafast, near real-time performance. Oracle says that it can offer less than 10 microseconds of latency between nodes, improving efficiency.

Oracle offers the widest range of bare metal GPU instances among major cloud providers, and scalability at any size up to 65,536 Hopper GPU clusters and 131,072 B200 GPU clusters, which are expected to come online in 2025. Oracle also offers very flexible VM instances, letting customers pay for only the capacity they need as they need it for any size workload, rather than offering fixed instance sizes.

With less overhead and fewer CPU cycles, RDMA helps Oracle offer its AI clusters at a lower cost: Oracle says it “consistently charges less than Amazon Web Services (AWS) for the equivalent compute capacity.”

Last night in the call, Oracle emphasized how cheap they are compared to the Big 3, stating: “We have gotten the entire Oracle Cloud, the whole thing, every feature, every function of the Oracle Cloud down to something we can put into a handful of racks, 3 racks, we call it Butterfly that cost $6 million. So we can give you a private version of the Oracle Cloud with every feature, every security feature, every function, everything we do for $6 million. I think the cost for the other hyperscalers is more than 100x that.”

The Importance of Vectorized Data

Oracle’s AI vector capabilities also stand out given Oracle’s database roots, offering native AI vector search capabilities with seamless integration to leading AI models from OpenAI, xAI, Meta, Cohere and more. AI vector search lets enterprises search both structured and unstructured data in a variety of manners, enabling intelligent, relevant and accurate AI responses utilizing their data. Oracle noted in Q3 that its Oracle Database 23ai can convert data into any vector format to be understood by an AI model of choice, facilitating AI training and inference on private data in Oracle’s Database.

The announcement of Oracle’s AI database is particularly interesting in terms of the stock extending its run. As explained in the call last night, the combination of vectorizing data to where it can be understood by AI models with the ability to connect private databases to AI reasoning models will result in enterprises unlocking higher value from AI. Here is what was

said: “Then we made it very easy for our customers to directly connect all their databases, all their new Oracle AI databases and cloud storage, OCI Cloud storage to the world's most advanced AI reasoning models, ChatGPT, Gemini, Grok, Llama, all of which are uniquely available in the Oracle Cloud. After you vectorize your data and link it to an LLM, the LLM of your choice, you can then ask any question you can think of. Who's offering that to customers? We'll be the first when we deliver it and demonstrate it at AI World next month.”

After a 35% Gap Up, is ORCL Stock Still a Buy?

After a historic day in the markets with Oracle up 36% after Tuesday's print, the main question that remains is if Oracle is still a buy or has the easy money been made? Considering that RPO is up 359%, cloud RPO is up nearly 500% and multi-cloud database growth is up over 1,500% – it’s well worth the time to look at what the technicals are saying as Oracle approaches a $1 trillion market cap.

Below, we discuss:

Is Oracle stock still a buy using technical analysis to discuss potential entry points.
What levels to watch to confirm Oracle still has room to run ahead of a highly anticipated Oracle Cloud World next month.
The key to Oracle’s stock expanding – which was not in the most recent earnings report, but rather, the two key items that will help drive the stock upward in 2026 and beyond.

The I/O Fund is a leading AI portfolio with 45% allocation to tech stocks going into 2023 before Wall Street caught onto the potential of AI, yet this stands at an 87% allocation to AI stocks today. With cumulative returns of 210%, we’d place #2 if we were a hedge fund and #5 if we were an ETF.

The Key to Oracle’s Stock Expanding is Two-Fold

There are a few key reasons Oracle stock can continue to expand:

1. AI Database:

Oracle is teasting a more beefed-up AI database, which management stated will officially launch at Oracle World Cloud next month, describing a combination of private enterprise data, large reasoning models and automated agents: “Who's offering that to customers? We'll be the first when we deliver it and demonstrate it at AI World next month.”

Oracle has already made major headway with AI embedded databases with 23ai, which converts vector data into contextual information. By connecting a database to Chat-GPT, there is more reasoning layered into the results.

HeatWave Gen AI is another piece to Oracle’s AI database offerings, as it combines systems for analyzing data. By combining systems, AI can run queries faster and cheaper compared to when data scientists have to use many different database systems. For example, HeatWave can theoretically replace a larger stack, such as AWS AI, Snowflake and Databricks, with the goal of lowering internal complexities for data engineers and also helps to lower costs.

The inference market will defined by size and quality of data for reasoning purposes, and Oracle sits on arguably the world’s largest enterprise data sets. Although we have grown used to compute driving the AI training market, there will be an important shift toward the data layer driving the inference market.

With Oracle embedding the AI database, inference will happen inside the database where the data resides. This is distinct from pulling data out of the database into the large language model, which is inefficient. Oracle’s move to embed the database supports a sustained, upward trajectory in the stock price.

Next month, Oracle is expected to expand its Oracle AI Database, which is designed to combine large language models – such as ChatGPT, Grok, or Gemini – with customers’ private databases and automated agents to help increase the depth of what models can achieve for enterprises. Oracle has already released many features for AI databases, yet management teased an announcement for next month’s Oracle World, hinting they will be offering more agentic AI.

Where Oracle is in a unique position is the company can keep private data secure, vectorize the data to be better understood by an AI model, while bundling the full-stack data layer with the public cloud and foundation models. It's a combination of lower costs, lower complexity and combining publicly available data with private data that is setting Oracle on a longer runway than what the market may be initially realizing – and yes, that statement includes the 36% move.

2. Look for Margins to Expand

If we assume the AI Database becomes a catalyst for Oracle, the likelihood of margin expansion is a key reason the stock could have more horsepower.

In the near term, Oracle is deliberately trading margin for growth as margins compressed QoQ across the board. Heavy restructuring and datacenter investments clipped profitability in the short term but underlying gross margin stability and YoY gains in adjusted income exhibit how Oracle is scaling Cloud while keeping structural profitability intact.

Longer term, the expectation is that margins should expand as the Company benefits from deploying AI databases across multicloud environments.

Here are the current operating metrics:

Q1 GAAP Gross Profit of $10.6B, down QoQ from $11.2B in Q4 but up compared to $9.4 in Q1’25. This represents a 71% gross margin, up from 70% in prior quarter and in line with 71% in prior year quarter.
Q1 GAAP operating income was $4.3 billion, down from $5.1 billion in Q4 but up from $4.0 billion in Q4’25. Q1 figures represent a 29% operating margin, down from 32% in prior quarter and 30% in prior year quarter.
Q1 Non-GAAP Operating income was $6.2 billion, down from $7.0 billion in Q4 but up from $5.7 billion reported in Q1’25. Q1 figures represent a 42% operating margin, down from 44% in prior quarter and down compared to 43% in prior year quarter.
This contraction is likely to resolve from a higher mix of AI database revenue – and that is where there could be more alpha for this stock
Q1 GAAP Net Income was $2.9 billion, down from $3.4 billion in Q4 and in line with $2.9 billion reported in Q1’25. Q1 figures represent a 20% net income margin, down from 22% reported in prior quarter and down from 22% in prior year quarter.
Q1 Non-GAAP Net income was $4.3 billion, down from $4.9 billion reported in Q4’25 but up from $4.0 billion reported in prior year quarter. This represents a 29% adjusted net income margin, down from 31% in Q4’25 and 30% in Q1’25.

What the Technicals Say about ORCL Stock:

The earnings gap today is the key driver for the current setup. When we see a vertical price move like the one following earnings, which is backed by maximum volume and momentum, it strongly suggests we are in the middle of a 3rd wave advance. This points to two bullish scenarios:

Green Scenario (Large 3rd Wave Breakout):

In this case, the rally from the April low into the July high represents wave 1. Today’s breakout would confirm wave 3, setting the stage for further upside with pullbacks along the way. The projected wave 5 targets fall in the $850 – $1000 range.

Blue Scenario (3-Wave Uptrend):
Here, the runoff the April lows was wave A, followed by a shallow B wave correction in July/August. The current move would represent wave C, with targets in the $394 – $567 zone.

Key Supports

For this trend to remain intact, the following levels should be monitored:

$306 – First warning level; a break below here raises doubts about higher prices, but does not invalidate anything.
$273 and $261 – Critical supports. As long as these levels hold, the broader uptrend remains valid, and the market should continue trending toward the target region.

Conclusion:

The I/O Fund initiated its position in Oracle because the inference market has not really kicked off yet. Some refer to this as early innings, but I would argue it's pregame. This conviction on timing is reinforced by Oracle’s latest earnings report, which showed a blowout in OCI growth and RPO, yet the current quarter fell short of expectations with a 6% QoQ decline. Even though Oracle fell shy of consensus estimates for the current quarter, the backlog and visibility point to robust AI-driven demand ahead. From a technical perspective, the stock also sets up for potential upside, creating what we see as a compelling risk/reward profile for investors.

The I/O Fund is a leading AI portfolio with 45% allocation to tech stocks going into 2023 – the highest allocation of any firm on record at the time, and today, this stands at an 87% allocation to AI stocks. With cumulative returns of 210%, we’d place #2 if we were a hedge fund and #5 if we were an ETF. Premium members receive real-time trade alerts, weekly webinars, deep dives on lesser-known AI stocks and more. Learn more hereLearn more here

Recommended Reading:

Nvidia Stock Forecast: The Path to $6 Trillion

Posted on September 5, 2025June 30, 2026 by io-fund

Two years ago, the April 2023 quarter delivered a historic 18% beat, followed by an even bigger 30% beat in July 2023. Compare that to the most recent quarter ending July 2025 — just a 4% beat, the smallest in two years. The narrowing beats combined with a rare QoQ decline in the Compute segment could be interpreted as Nvidia is running out of steam.

I am frequently asked by notable media anchors if Nvidia could be topping, and as you’ll see below, I substantiate why Nvidia is not running out of steam. I spoke with Charles Payne from Fox Business and Caroline Hyde from Bloomberg, followed by a 45-minute interview on the topic with Maggie Lake of Wealthion. In these conversations, we discussed some of the most pressing issues that AI investors face – does China revenue matter for Nvidia, how does the world’s most valuable company continue to grow, are we in an AI bubble, and lastly – where is the I/O Fund positioning next.

For those new to my firm, the I/O Fund is among the top-performing AI portfolios globally. Our track record was built in part by taking a significant Nvidia position years before the market embraced the AI story, with allocations as high as 10–20%. While our calls on Nvidia are well recognized, we’ve also held a number of lesser-known AI winners that together have driven a cumulative return of 210%. To put that in perspective, if we were a hedge fund, we’d rank #2 — and if we were an ETF, we’d rank #5.

What that ranking should communicate to you is that the I/O Fund does not rest on our laurels, we rigorously pursue a portfolio positioning that will outperform. Let me rephrase this to say that my firm will not continue to hold Nvidia out of fandom, rather, we will continue to hold Nvidia only if we believe the stock continues to offer alpha.

Below, I explain why this stock has the potential to outperform the indexes — and how we’re positioning to capture even greater returns from lesser-known stocks that are riding Nvidia’s momentum.

Is There Alpha Left in Nvidia (NVDA) Stock?

After being crowned the world’s most valuable company with a market cap over $4 trillion, it's natural to wonder if the easy money has already been made. For investors, the debate comes down to whether Nvidia can keep outpacing both the Nasdaq and its semiconductor peers given its phenomenal run.

Regardless of market fluctuations, Nvidia’s product road map is not slowing down. As described in the videos below, Nvidia is on “the eve of releasing its next generation of GPUs” combined with its enviable software platform CUDA and strong QoQ growth in AI networking (NVLink, Spectrum-X, InfiniBand).

In the interview below with Charles Payne of Fox Business, which took place ahead of earnings, I offer key reasons that Nvidia continues to offer alpha as we head into calendar year 2026 and why we plan to buy the stock on any dips.

Yesterday, I discussed one specific reason that I’d be a buyer of Nvidia $NVDA if the stock sold off due to China. @cvpayne and I also talked about how to position given some AI stocks seem to be in a bubble (while others are not). pic.twitter.com/uNMX2svK8l

— Beth Kindig (@Beth_Kindig) August 28, 2025

I also had the opportunity to talk to Maggie Lake of Wealthion on the importance of Nvidia and why the company is moving from its 2.0 era to the 3.0 era, and how this transition from server-scale to a rack-scale AI systems company is key as to why Nvidia has further room to run. In the clip, I also describe the opportunity that Nvidia is poised to capture that will be as large as the AI hardware opportunity.

The AI Bubble: Fact or Fiction

Fears as to whether AI is reaching bubble territory were ignited this past month when Sam Altman compared AI to the dot-com era, stating: “When bubbles happen, smart people get overexcited about a kernel of truth. If you look at most of the bubbles in history, like the tech bubble, there was a real thing. Tech was really important. The internet was a really big deal. People got overexcited.”

Given OpenAI is seeing about $20 billion in revenue as of now with no profits, it makes sense that Altman would be concerned about valuations. However, a lack of profits is certainly not Nvidia’s concern. The company had a GAAP operating margin of 60.8% in Q2 and an adjusted operating margin of 64.5% with over $13 billion in cash flow for the quarter.

There is clearly a bifurcation in the AI economy as some companies must spend heavily to compete, while others are profiting heavily from the steep competition.

In as lucid of a manner as possible, I break down in the videos below the following:

Why technically every tech trend goes through a boom/bust, why this should not scare investors off, plus the strategy investors can use to successfully win with surging tech trends
What companies are more insulated from the effects of a bubble and why AI software like OpenAI is at higher risk
The differences between an enterprise technology and a consumer technology in terms of how AI will repay Big Tech capex

Nvidia’s Stock Can Reach $6 Trillion Market Cap by Next Year

In an interview that went viral with Caroline Hyde of Bloomberg, I discussed why it’s important to not focus too closely on quarterly earnings reports for determining Nvidia’s growth potential, and to rather focus on GPU generations.

If we look at the revenue potential for quarterly data center revenue once Blackwell and Blackwell Ultra ship in volume, this leads us to a $6 trillion valuation by next year.

Predicting 50% upside may not seem like much given Nvidia is up 325% since early 2024 and up 4,000%+ since my firm first bought the stock in 2018 – however, there is a chance investors have an opportunity to get Nvidia lower.

For example, as recent as January of this year, my firm predicted Nvidia would dip below $100 – this was published before the DeepSeek news hit. This afforded our followers a better entry, leading to higher upside. Next week, we will be updating our buy plan for Nvidia for free. Don’t miss it by signing up for our weekly newsletter here.

Nvidia $NVDA is at roughly a $160B run rate for the data center.

In an interview with Caroline Hyde from @technology, I discuss the path to a $300B run rate plus the constraints that Nvidia and the broader AI market face as we approach future generations of GPUs.… pic.twitter.com/CzzYvKpC7x

— Beth Kindig (@Beth_Kindig) August 27, 2025

The Inevitable AI Trend that will Profit Most from Nvidia’s Upcoming Generation of GPUs

Nvidia 1.0 was defined by the company's lead in gaming, whereas Nvidia 2.0 was defined by the introduction of Tensor Cores with Volta. These cores evolved through Ampere and the Hopper generation, ultimately leading to the transformer engine and more precision support for accelerating AI compute. The market eventually caught onto the enormity of the AI market as Hopper facilitated more accurate GenAI applications coming to market. You can read more about this history in the analysis: “Here’s Why Nvidia Will Reach a $10 Trillion Market Cap by 2030.”

Looking forward, what I am dubbing as “Nvidia 3.0” will be the shift from 8-GPU server-scale systems to 72-GPU rack-scale systems — which is augmented with the back-to-back release of Blackwell and Blackwell Ultra. Nvidia’s new rack-scale systems will mark one of the most significant turning points in its history. Regardless of what market participants think about Big Tech’s surging capex spend, the spend will continue to rise as those who get the NVL72 systems (that are beginning to ramp now) will have a critical advantage over those still on the HGX or DGX systems with 8 GPUs.

There is one inevitable trend that will profit most from the upcoming generation of GPUs, with significant momentum from Nvidia’s new rack-system architectures. This 5X to 9X opportunity that is already beginning to take shape with material evidence of the momentum.

Don’t miss out on what a leading AI portfolio and analyst team is saying about where to position next to take advantage of AI’s next major move.

AI Networking is the #1 Opportunity from Nvidia’s Upcoming GPUs

Under the hood, what defines the upcoming generation of GPUs from the previous generation is AI networking. When you move from 8 GPU severs to 72 GPU rack-scale systems, your intra-rack networking shifts as does the rack-to-rack for larger clusters due to higher GPU density. For example, we’ve seen hyperscalers plan to deploy 1 million GPU clusters by 2027-2028 up from clusters with 10s of thousands right now. Therefore, both scale-up and scale-out architecture are going through a rapid transformation on the networking side.

In the video below, I discuss why the I/O Fund is concentrated in AI networking stocks as our #1 trend right now:

Conclusion:

Nvidia and its ecosystem of suppliers are the best way to position for the rest of 2025. While headlines were fixated on China, the bigger story is clear: H20s contributed about $4B, while Blackwell generated an estimated $28B this past quarter — over a $100B run rate. This is why my stance has been to focus on GPU generations to track the AI opportunity rather than quarterly earnings or any single geographic area for seven years and counting. Semiconductors are particularly tricky as the cyclical bottom can often be the best time to buy.

Keep an eye out for the I/O Fund's new buy plan for Nvidia, hitting inboxes next week. Our last buy plan provided an entry under $100 when the market felt the stock was invincible. This led to entries with real-time trade alerts at $94.48 and $87.99 with real-time trade alerts sent to our Premium Members.

For 2025, the I/O Fund has worked to identify key Nvidia suppliers with Blackwell on deck to ramp significantly, sharing our in-depth research on stocks within the AI networking stack. We hold high allocations in two specific Nvidia AI networking suppliers.

Sign up to join our upcoming webinar, held every Thursday at 4:30 pm EST, where we discuss buy zones for AI stocks and more. Learn more hereLearn more here

Recommended Reading:

Broadcom Hints of AI Revenue Growth Accelerating in FY26; Backlog of $110B

Posted on September 5, 2025June 30, 2026 by io-fund

Broadcom Q3 results confirm the stock as the number two player in AI as a combination of AI networking, custom silicon and AI software propelled total revenue to record levels. Semiconductor Solutions accelerated nine points to 26% YoY growth due to a rebound in AI accelerators (+63% YoY) and networking (170% YoY). While non-AI Semiconductor revenue remains weak and flat sequentially, VMware’s contribution helped prop up the segment.

Looking ahead, Broadcom’s Q4 guide implies acceleration into year-end, with total revenue expected to reach $17.4B. More importantly, we picked up on a subtle shift in commentary from CEO Hock Tan regarding AI revenue growth in FY26, which should place Broadcom firmly above $30 billion for next fiscal year. Tan stated: “[…] we now expect the outlook for fiscal 2026 AI revenue to improve significantly from what we had indicated last quarter.”

As we previously discussed this summer, demand for inference is booming. Broadcom’s edge goes beyond the fact that custom accelerators are often multiples cheaper than Nvidia’s GPUs for inference tasks – it's that custom silicon is increasingly performant with each generation. By optimizing algorithms (software), Big Tech can drive higher performance from large language models (LLMs) — which helps to drive down costs while also increasing output for specific workloads. For example, a rough idea as to how much it costs Nvidia to make merchant GPUs is estimated around $3,000 to $5,000 whereas the company charges $25,000 to $30,000 – hence the AI leader’s excellent margins. Reducing Nvidia’s high pricing power is what Big Tech is after and this can be accomplished both in the hardware costs but also through optimizing the workloads for specific use cases.

Big Tech is prominent in Broadcom’s custom silicon customer list, which includes Google and Meta. ByteDance reportedly emerged as the third customer last summer. Tonight, a fourth customer was announced for a $10 billion XPU order, which was likely either OpenAI or perhaps Apple, as both were heavily rumored to be prospective customers (note: customer name was not offiically confirmed)

“Last quarter, one of these prospects release production orders to Broadcom. And we have accordingly characterized them as a qualified customer for XPUs. And, in fact, has secured over $10 billion of orders of AI regs based on our XPUs.”

Perhaps most interesting, Hock Tan slipped into the opening remarks that their backlog is at $110 billion. I was glad when an analyst circled back to this comment as this backlog is astonishing, to say the least.

More details are below!

170% AI Networking Growth – Outpacing Street Models

Q3’25 Revenue was $15.95 billion, beating estimates for $15.82 billion, and reflecting top line growth of 22.0% YoY and 6.3% QoQ. This represents record quarterly revenue, driven by custom AI accelerators, networking switches, and VMware strength in Infrastructure Software.

Looking ahead, management provided Q4’25 guidance of $17.4 billion of revenue, implying 24% YoY growth and a slight uptick to 9% QoQ growth. This guide suggests that AI strength will more than offset flat non-AI revenue, while VMware remains a stable contributor.

Key Segments

Semiconductors were the main growth driver, fueled by AI accelerators and Ethernet networking. Infrastructure Software, which includes VMware acquisition, remained strong even though growth decelerated slightly when compared to prior quarters,

Semiconductor Solutions Revenue accelerated nine points to 26% YoY at $9.17 billion. AI Semiconductor revenue surged 63% YoY to $5.2B, showing re-acceleration after a slower Q2 (+46% YoY). AI now represents 57% of Semiconductor revenue and 32% of total company revenue.

Management guided Q4 AI revenue to $6.2B, which would represent ~19% sequential growth and eleven consecutive quarters of YoY growth. Though Broadcom did not lay out a FY25 AI revenue target, Q4 implies Broadcom is guiding for $19.9 billion in AI revenue for the year, up 63% YoY from $12.2 billion in FY24.

Non-AI Semiconductor revenue remained flat at ~$4B, indicating continued end-market softness outside of AI. As seen below, the gap between AI and non-AI revenue is widening as AI growth accelerates.

Infrastructure Software Revenue of $6.8B, ahead of guidance of $6.7B, up $190M QoQ from $6.6B in Q2’25 and up from $5.8B in Q3’24. $6.7B in revenue represents 43% of total revenue and 17% YoY Growth.

Operating Leverage Offsets Gross Margin Compression

Broadcom delivered balanced profitability in Q3 FY25. While AI hardware growth slightly compressed GAAP and Non-GAAP gross margins, the Company leveraged its scale and VMware’s recurring revenue to expand operating and net margins YoY. Broadcom’s ability to maintain a 67% EBITDA margin during a period of rapid AI expansion demonstrates exceptional cost control and pricing power, positioning it well for the Q4 acceleration. See below for in-depth breakdown of GAAP and Non-GAAP margin figures:

GAAP gross margin of 67.1%, down 90 bps QoQ from 68% in Q2, and up from 63.9% in the same period last year. The margins were down sequentially due to custom silicon mix, which typically carries lower margins. Broadcom is flexing its ability to manage cost structures even as AI accelerators scale rapidly.

Non-GAAP gross margin of 78.4%, down 100 bps QoQ from 79.4% in Q2 and up 100 bps YoY. This is driven by the increase in AI hardware revenue, which has lower gross margins compared to VMware’s high-margin software business.

GAAP operating margin of 36.9%, down 190 bps QoQ from 38.8% in Q2, but up significantly from 29% in the same period last year. QoQ decline reflects seasonal opex and YoY improvement reflects strong operating leverage.

Non-GAAP operating margin of 65.5%, slightly up QoQ from 65.3% in Q2’25 and up 180 bps from 63.7% in Q3’24. Broadcom is maintaining tight opex control even as it invests heavily in AI and networking. Non-GAAP margin stability reflects strong execution in managing both segments.

GAAP Net Margin of 26.0%, down from 33.1% in Q2’25 and up from (14.3%) in Q3’24. Non-GAAP Net Margin of 52.7%, up from 51.9% in Q2’25 and up from 43.6% in Q3’24. Adjusted profitability is expanding as Broadcom benefits from both VMware’s recurring software revenue and AI-driven scale.

Adjusted EBITDA of $10.7B, compared to $10.0B in Q2’25 and $8.2B in Q3’24. This represents an Adjusted EBITDA margin of 67.1%, compared to 66.7% in Q2’25 and 62.9% in Q3’24.

Key Takeaways:

AI Mix Impact on GM: As AI Semi revenue grows (now sits at 32% of company revenue) there is slight gross margin pressure because custom silicon and hardware products carry lower margins.
Offset by Operating Leverage: Despite gross margin compression, non-GAAP operating margin expanded YoY thanks to scale and cost efficiencies.
Stable EBITDA Margin: Maintaining a 67% EBITDA margin while growing 22% YoY highlights Broadcom’s unique profitability profile when compared to peers.
Watch Forward Trend: With AI projected to reach $6.2B in Q4, mix shift could continue to pressure GAAP gross margins further. Look for growth from VMware and networking to keep non-GAAP margins steady or even up.

EPS Growth driven by Top-Line Growth & Economies of Scale

Non-GAAP EPS growth is the real story here: up 38% YoY, reflecting Broadcom’s ability to capitalize on AI strength with containing costs. GAAP EPS lagged slightly due to accounting for VMware’s amortization and elevated interest expense – not driven by any core business weakness.

EPS growth of 36.3% outpaced revenue growth of 22%, highlighting the company’s dual-engine model: AI hardware scale drives top-line growth while VMware’s recurring software offsets margin volatility and lifts EPS. Investors will appreciate the strong non-GAAP EPS beat and expansion in operating margins, especially given Broadcom’s premium valuation, where both EPS stability and growth are critical to sustaining multiples.

GAAP EPS of $1.02, compared to $1.03 in prior quarter and ($0.40) in prior-year quarter.
Non-GAAP EPS of $1.72, compared to $1.58 in prior quarter and $1.24 in prior-year quarter.

Record Free Cash Flow Conversion

Broadcom converted 44% of revenue into free cash flow, placing it among the top-tier semiconductor companies for cash generation. Strong OCF growth driven by AI semiconductor momentum and VMware’s software contribution. Broadcom continues to de-risk its balance sheet, reducing debt by $3B sequentially while building cash reserves and modestly increasing inventory ahead of anticipated Q4 AI revenue acceleration. With high FCF conversion and a strong balance sheet, Broadcom should have ample flexibility to return capital to shareholders and fund future AI growth initiatives, reinforcing its premium valuation. See below for an breakdown on cash flow figures:

GAAP operating cash flow of $7.2B, up 10% QoQ compared to $6.5B in Q2’25 and 57% YoY $4.6B in Q3’24. This represents an operating cash flow margin of 44.9%, up from 43.7% in Q2’25 and 36.7% in Q3’24.
Free Cash Flow of $7.0B, compared to $6.41B in Q2’25 and $4.5B in Q3’24. This represents a free cash flow margin of 44.0%, compared to 42.7% in Q2’25 and 35.6% in Q3’24.
Cash and Cash Equivalents of $10.7B, up from $9.5B as of Q2’25 and up from $10.0B as of Q3’24.
Debt of $64.2B, down from $67.2B in Q2’25, and down from $69.9B in Q3’24.
Inventory of $2.2B, up from $2.0B in Q2’25, and up from $1.8B in Q3’24.

Valuation

Broadcom currently trades at a premium to Nvidia, sitting at 19x forward revenue vs Nvidia 17.6x. Based on forward earnings, AVGO trades at 38x forward earnings, 13% above Nvidia and 18% above semi-industry average.

This premium suggests that the market is pricing in 70%+ AI revenue CAGR through FY27, above management’s current 60%. The implication here: Any sign of AI growth slowing or continued margin pressure could trigger a valuation reset.

Earnings Call Q&A

Hock Tan Hints that Strong AI Growth is Incoming

Last quarter, Hock Tan had stated the following: “And reflecting this, we may actually see an acceleration of XPU demand into the back half of 2026 to meet urgent demand for inference on top of the demand we have indicated from training. And accordingly, we do anticipate now our fiscal 2025 growth rate of AI semiconductor revenue to sustain into fiscal 2026."

This evening, the following was stated:

“we now expect the outlook for fiscal 2026 AI revenue to improve significantly from what we had indicated last quarter.”

When asked later to clarify, Tan reaffirmed our understanding that it would mean above 60% growth and it was also stated the majority of the growth would come from XPUs with networking’s share of AI revenue declining next year due to XPU strength.

“Let's answer the first part first, if I could be so bold as to suggest to you, when I — last quarter when they said, "Hey, the trend of growth of '26 will mirror that of '25which is 50%, 60% year-on-year. That's really all I said. I didn't — but of course, it comes up 50%, 60% because that's what '25 is.

All I'm saying, if you want to put another way of looking at what I'm saying, which is perhaps more accurate is we're seeing — the growth rate accelerates as opposed to just remain steady at that 50%, 60%. We are expecting and seeing 2026 to accelerate more than the growth rate we see in '25. And I know you love me to throw in the number at you, but we are not supposed to be giving you a forecast for '26. But best way to describe it, it will be fairly material improvement.”

Back of the napkin math:

Given that Hock Tan disclosed that Broadcom had secured $10B+ of orders related to the new fourth customer, current growth projections may materialize far too low. We have some estimates from HSBC placing just ASICs revenue at $28.3B (+128% YoY) with only $2.5B contribution from customers outside GOOG and META.

The Street is at $19.9B for ASICs in FY26; assuming the Street is still at $28-29B AI revenue for FY26, this places Networking at $8-9B.

However, assuming XPU strength and strong networking demand (as seen throughout the ecosystem) can drive an acceleration to 70% YoY in FY26, this projects $33.8B in AI revenue up from $19.9B estimated in FY25 given the implied guide.

This ramp is supported by estimated ~80% increase in CoWoS allocation from ~83K in FY25 to ~150K in FY26, after Broadcom reportedly increased orders last week from 120-125K expected in FY26.

Notably, Hock Tan did say the additional $10 billion would be recognized in Q3 of FY2026 during the call.

$110 Billion Backlog

Broadcom stating a $110 billion backlog was not on my Bingo card tonight. Wow! That is certainly a strong statement in terms of Broadcom’s prospects for continued AI growth. It’ll take me a day or two to process what that could mean for the next 1-3 years, assuming some of this is supply constrained.

An analyst did ask for clarity and since it’s a rather large number to provide, especially since it’s primarily AI-driven, I’m quoting it in full for you below:

Stacy Rasgon. Bernstein & Co.

I was wondering if you could parse out this $110 billion backlog. Did I hear that number right? Could you give some color on the makeup of it — like how far does that go — and like how much of that $110 billion is AI versus non-AI versus software?

Kirsten Spears CFO:

Well, yes, Stacy, we generally don't break up back on digital to give you a sense of how strong the business is as a whole for the company, and it's largely driven buying AI in terms of growth. Software continued to add on a steady basis. And non-AI, as I indicated, has grown double digits. Nothing compared to AI, which has grown very strongly. Give you a sense, perhaps fully 50% of it at least is semiconductors.

Stacy Rasgon, Bernstein:

Okay. And it's fair to say that semiconductor piece, it's going to be much more AI than non AI.

Hock Tan, President:

Right.

Custom Silicon is Progressively Gaining Market Share

During the discussion, an analyst asserted that custom silicon could surpass GPUs in terms of market share. That’s a tall order given Nvidia will typically be 1-2 years ahead of any custom programs (thereby offering an advantage to those who remain with their GPUs) and the two companies will likely end the year nearly $180B apart in AI revenue with AVGO around $20B and NVDA over $200B (could see $212B).

However, I’ve also argued inference will provide an opening for Broadcom and AMD to meaningfully compete on AI accelerators. Therefore, I’m all ears and we will be watching this closely as we move along 2026-2028.

Harsh Kumar, Piper Sandler:

Hock, congratulations on all the exciting AI metrics and thanks for everything you do for Broadcom and sticking around. Hock, my question is, you've got 3 to 4 existing customers that are ramping. As the data centers for AI clusters get bigger and bigger, it makes sense to have differentiation, efficiency, et cetera, therefore, the case for XPUs. Why should I not think that your XPU share at these 3 or 4 customers that are existing will be bigger than the GPU share in the longer term?

Hock Tan, CEO:

It will be. It's a logical conclusion, Harsh, you're correct. And we are seeing that step by step. As I say, it's a journey. It's a multiyear journey because it's multigenerational, because these XPUs don't stay still either. I'm doing multiple versions, at least 2 versions, 2 generation versions, for each of these customers we have. And with each newer generation, they increase the consumption, the usage of the XPU. As they gain confidence, as the model improves, they deploy it even more. […] And that's why I say we progressively gained share.”

Conclusion:

Broadcom’s AI networking growth of 170% YoY and sharp rebound in custom accelerators at 63% YoY was certainly a highlight. In addition to strong AI growth, the company maintained a 67% EBITDA margin despite lower-margin hardware scaling, while record free cash flow of $7.0B highlighting the company’s financial strength. With AI now making up a third of total revenue and Q4 revenue projected at $6.2B, Broadcom is slated to accelerate XPUs as a primary beneficiary of the inference era.

The nod toward accelerating growth in 2026 is why the stock went from flat to +5% AH, as the CEO seemed to hint that 70% AI growth or higher is not out of the question. Essentially, there are still four months to go in 2025 and Broadcom is gearing up for a strong 2026 already. The backlog of $110 billion was a “sit up in your seat” moment as it will force analysts to ponder – just how long will it take to work through that backlog? One can safely assume the backlog will only grow from here, as Broadcom is communicating they are preparing to be a strong contender to Nvidia toward the end of this decade.

That’s a wrap! The I/O Fund’s earnings season is officially coming to a close. Keep an eye out for deep dives over the coming weeks plus coverage of notable earnings reports on stocks we don’t own.

Recommended Reading:

Credo’s Hypergrowth Intact as 274% YoY Growth Leads to Massive Bottom Line Expansion

Posted on September 4, 2025June 30, 2026 by io-fund

Credo easily had one of the best earnings reports across the tech sector this quarter. The company reported growth of 274% YoY and 31% QoQ for revenue of $223.1 million. This beat estimates on the top line by 17% with management raising full-year revenue growth outlook by 35 points, from 85% YoY to 120% YoY. This corresponds to revenue of at least $960 million, and considering Credo expects two new hyperscaler customers to ramp in mid to late FY26, the company could well be on a trajectory to reach $1 billion in revenue this fiscal year. This would be an entire year ahead of analyst expectations, which projected $1 billion in revenue in FY27 heading into the report.

The bottom line also shined with adjusted EPS beating estimates by 44.4%. This represents growth of 1,200% YoY from a thin $0.04 in the prior-year quarter. Triple-digit growth of 425% on the bottom line is expected to follow although flat QoQ.

To be prudent, Credo’s Q2’s guide could be perceived as soft, at ~5.4% QoQ versus the 31% QoQ reported in Q1. Additionally, analysts were concerned about lumpiness in lead customers, although to be fair, Credo has done quite well with few customers up to this point by tackling a large total addressable market (TAM) with reliable yet cost-effective solutions. Credo’s proprietary serializer/deserialzer (Ser/Des) technology, active electric cables and digital signal processing (DSPs) are the cornerstone of its IP portfolio, giving the company a significant competitive advantage as it enables the power-efficient connectivity and reasonable pricing that the company is known for.

In a nutshell, this is why Credo is reporting surging growth in a highly competitive market: “Reliability and power efficiency [leads] to choosing AECs over optical solutions as they are up to 1,000x more reliable and consume half the power. AECs virtually eliminate link fabs, which are intermittent losses of connection, boosting cluster reliability and productivity while reducing power consumption.”

More information on how Credo plans to sustain growth into 2026 is noted below in the Q&A section, along with more details from tonight’s standout earnings report.

Revenue Growth Accelerates Nearly 100 Points

Credo smashed its own guidance by more than $33 million as it reported $223.1 million in revenue in Q1, riding strong demand for its power-efficient high-speed connectivity solutions. Growth accelerated nearly 100 points sequentially to 274% YoY, well ahead of estimates for 219%. On a sequential basis, revenue rose 31% QoQ, a slight acceleration from 26% QoQ in fiscal Q4.

Coming into the report, the bar was already set high as estimates called for a 150-point revenue acceleration over three quarters to >200% YoY growth.

For fiscal Q2, Credo guided for $230 to $240 million, or 226% YoY at midpoint, decelerating nearly 50 points from Q1. Sequentially, the guide feels more conservative at ~5.4% QoQ considering the product momentum Credo has behind it with AECs, optics and DSPs.

As of Q4, management had provided initial FY26 revenue guidance of >$800 million representing 85% YoY growth. Now, this has been updated to growth of 120% YoY, or north of $960 million for the year.

Customer Concentration:

In Q4’25, Credo disclosed its largest customer accounted for 61% of revenue (widely believed to be Microsoft). Management said they expected up to five >10% customers in FY26, up from three in FY25. Two additional hyperscalers were highlighted as ramping in mid-and late-FY26, with potential to become > 10% customers.

In its Q1’26, Credo provided some additional color around progress on these fronts. Management confirmed that the three >10% customers noted in Q4’25 were consistent in Q1’26, coming in at 35%, 33% and 20% of total revenue respectively.

The largest customer in FY25 continues to be the largest customer so far in FY26, signaling deepening relationships on-top of these new customer adds. Management also noted that Q1 included its first material revenue contribution from a 4th hyperscaler, with the 4th customer expected to surpass the >10% threshold by year end. Management expects continued progress around diversification through FY26.

Key Segments

Product Revenue: came in $217.1M, up 279% YoY and 31% QoQ. This is just shy of +303% YoY growth in Q4FY25 but still represents hypergrowth scale. Credo’s core engine remains AECs, which continue to benefit from rack-scale AI deployments). The fact that product revenue sustained triple-digit YoY growth while already running at a $200M+ quarterly pace suggest demand remains well ahead of consensus estimates.

IP License Revenue of $6.0M vs $4.2M in Q4 represented 44% QoQ. Still a smaller slice of the “revenue” pie but showing sequential growth. While licensing is still <3% of revenue but provides margin-accretive diversification. Growth here reinforces the stickiness of Credo’s SerDes IP, but the story remains dominated by physical product sales.

Engineering Services not broken out in this release, likely immaterial versus Product Sales.

Key Takeaways:

AEC remains the primary growth driver – scaling with hyperscaler AI deployments. Product revenue is highly concentrated.
Optics / DSP not broken out numerically this quarter, but management previously guided for 100% growth in FY26. Given the revenue beat, optics may already by contributing to incremental upside.
Retimers / PCIe 6 are still in early stage, but momentum in design wins should show up later FY26- FY27.

Operating Margin Shows Strong Sequential Expansion

GAAP Gross Margin was 67.4%, up from 67.2% last quarter and up from 62.5% in prior-year quarter. Adjusted gross Margin was 67.5%, down from 68.0% last quarter but up from 62.9% in prior-year quarter. Coming into the report, Credo had guided for: GAAP gross margin of 63.4 – 65.4% and adjusted gross margin of 64 – 66%. This shows continued expansion despite hypergrowth, a rare feat at this scale. Gross margins are expanding as volumes surge – evidence that AEC and optics pricing power is intact, and scale is not being bought at the expense of margin.

GAAP Operating Margin was 27.2%, up from 19.9% last quarter and up from (24.2%) in prior-year quarter. Adjusted Operating Margin was 43.1%, up from 36.8% last quarter and up from 3.7% in prior-year quarter. This is the standout – massive operating leverage as opex grew only ~11% QoQ vs. the 31% pick up in revenue. Non-GAAP operating margin of 40% signals that Credo is already functioning with elite efficiency, while still in hypergrowth mode.

GAAP Net Margin was 28.4%, up from 21.5% last quarter and up from (15.9%) in prior-year quarter. Adjusted Net Margin was 44.1%, up from 38.4% last quarter and up from 11.7% in prior-year quarter. Net Margins largely mirror the operating leverage story – Credo has become a profit machine far earlier in its lifecycle than most hardware names. The non-GAAP margin profile rivals leading semiconductor companies, while GAAP remains strong despite rising stock comp.

Adjusted EPS up 1,200% YoY

Credo reported GAAP EPS of $0.34, up from $0.20 last quarter and up from ($0.06) in prior-year quarter.

Q1’s adjusted EPS was $0.52, up 1,200% YoY from a thin $0.04 in the prior-year quarter and beating estimates by more than 44%. This also was a ~50% sequential improvement from $0.35 in Q4 driven by strong margin expansion down the line.

Credo is currently expected to report 435% YoY growth in adjusted EPS in Q2 to $0.37 and 58.4% growth in Q3 to $0.40, though given the margin strengths and sizable Q1 beat, these figures could move higher in the coming days.

Solid Cash Flow Generation in Q1

Credo’s balance sheet and cash flow position remain a core strength, underpinned by solid profitability, disciplined working capital management, and a debt-free structure. The company exited Q1 FY26 with $480 million in cash and investments, up from $431 million last quarter which should provide ample liquidity to support on-going product ramps and elevated R&D spend. Operating and free cash flow remained robust despite a sequential moderation from Q4’s unusually strong collections, with improvements in receivables management (DSO down to 73 from 86) helping offset continued investment in inventory. Inventory days held steady even as dollar levels rose, suggesting stocking is keeping pace with sales growth rather than accelerating further. Payables contracted notably, with DPO falling to 68 days from 91, reflecting less supplier financing and contributing to a softer cash conversion cycle. Overall, Credo continues to generate healthy cash margins, maintain a fortress balance sheet, and reinvestment modestly in capacity through capex.

Though cash flows moderated slightly from Q4, both operating and free cash flow margins remained >20%. This represented substantial YoY expansion in OCF and FCF margins of 35 to 45 points.

GAAP Operating Cash Flow of $54.2 million, down slightly from $57.8 million last quarter and up from ($7.2 million) in the prior-year quarter. This represents an OCF Margin of 24.3%, down from 34.3% last quarter but up significantly from (12.1%) in the prior-year quarter.
Free Cash Flow of $53.1 million, down from $54.2 million last quarter and up from ($13.1 million) in the prior-year quarter. This represents an FCF Margin of 23.8%, down from 31.9% last quarter but up more than 45 points from (21.9%) in the prior-year quarter.
Cash and Cash Equivalents (including short-term investments) of $479.6 million, up from $431.3 million last quarter. Credo remains debt-free.
Accounts receivable of $181.2M up from $162M in Q4’25 and $71.8M in Q1’25. Q1’26 implies DSO of 73 days compared to 86 days as of Q4’25, reflecting a meaningful improvement in collections. Credo is monetizing sales faster despite rapid growth which is supportive for cash flow sustainability and offsets some of the working capital drag from inventory.
Inventory of $116.6M, up from $90.0M last quarter and up from $31.5M in prior-year quarter. Q1’26 DIO (Days Inventory Outstanding) of 144 days is largely flat compared to Q4’25 DIO of 145 days. Inventory levels are holding steady relative to COGS, despite inventories increasing in dollar terms. This suggests the big build last quarter may have been a step-function while Q1 was more consistent stocking in line with higher sales volume.
Accounts payable of $54.9M, down from $56.2M in prior quarter and $38.47 in prior year. DPO (Days Payable Outstanding) in Q1’26 is down to 68 days compared to 91 days in Q4’25. This reflects a notable decline of 23 days in payables days and could be due to earlier supplier payments, changes in terms, or timing effects.
Capex was $2.8M, driven mostly by production equipment. This figure is down from $3.7M in prior quarter and down from $22.0M in prior year quarter. production equipment

Earnings Call Q&A Highlights

Commentary on Customer Concentration:

There was a moment during the call when the price action was more muted +5% versus +12% now. I believe it happened when management disclosed the lead customer represented 35% of revenue in Q1 compared to 61% last quarter. However, the market is being fickle if so, as customer diversification should be seen as a strength.

Here is what the CFO stated:

“Vivek Arya BofA Securities:

First set of questions is on the AEC market. If you could quantify how large each of your 10% customers were if they were the same as you had in the prior quarter?

And then Bill, if I zoom out, the market for you is now run rating closer to $1 billion or so. How large do you think this market is over time? And do you think this market is cannibalizing traditional copper? Or do you think at some point, it is even cannibalizing optical transceivers?

Daniel Fleming, CFO:

Vivek, this is Dan. Let me address the 10% customer question that you had. So as we mentioned in our prepared remarks, we had three 10% customers in Q1. They were the same three customers that were 10% customers in Q4. The mix was a little bit different, though. Our largest customer was 35% of revenue. Second largest was 33% and the third largest was 20%. So we're quite pleased with that kind of the customer diversity that we demonstrated within Q1.

But having said that, we expect continued diversification, as Bill highlighted, throughout fiscal '26. We do see two additional hyperscalers ramping, one of which, which was our fourth hyperscaler, we mentioned should reach to be a 10% customer for the full year of fiscal '26.

And then the last thing I'll mention on customer our largest customer for fiscal '25 is the largest driver of our growth in fiscal '26 as we stand right now and look forward. So that's an important factor to bear in mind as well as you look at how our year will progress..”

Later, an analyst asked for more clarity as to why the lead customer was down yet management stated the largest customer from last year will remain their largest customer for FY26 “by far, actually.”

Copper AECs versus Optical:

I think it’s worth repeating why Credo is seeing outsized demand in an otherwise crowded market – and one that can change rapidly in terms of which suppliers are confirmed, and if direct active cables are used (DACs), active electric cables (AECs) or optional solutions.

The question was: “And do you think this market is cannibalizing traditional copper? Or do you think at some point, it is even cannibalizing optical transceivers?”

This question is important as it’s asking why can Credo take market share with AECs from both traditional copper and optical.

The CEO stated: “But I think with the emphasis on reliability as it relates to clusters that we see customers really considering using AECs. It's really driven by reliability. I mentioned on the prepared comments that we're up to 1,000x more reliable, effectively reducing length laps and having the uptime of the cluster being much more.

Again, reiterating that if you have got a single link flap in, say, 10,000 or 100,000 or 1 million GPU cluster, it brings the entire cluster down because there's no redundancy from that NIC to tour connection. And so we're actually seeing the TAM expanding. And I think for the first time in history that you're seeing copper replacing optical connections. So we're quite bullish on the market generally.”

Specifically, where Credo has done well is with row-scale connectivity, which links racks in a single row. However, there are additional ongoing opportunities for Credo in rack-scale networking with NiC and Tor (top of rack) switches, as these both represent distances ideal for the reliability and power efficiency of AECs compared to optical. NiC-to-Tor would be a new opportunity for Credo: “Again, to reiterate, we see a huge opportunity from NIC to TOR applications as well as switch rack applications. And that is for front end and really emphasized much more strongly in back end, both in scale out and scale up.”

Overall management felt confident that the advantages AECs offer will continue to see increased adoption by hyperscalers with scale-out being the main driver now for Credo, yet scale-up being a future driver as well that is “an order of magnitude larger”

“When we talk about rack-to-rack, this is really the expanding part of the TAM really in both markets from the standpoint of AI back-end networks. as well as switch racks. Switch racks are starting to go to multiple rack architectures. And so as we talk about the near-term opportunity for rack-to-rack, it's really represented by the scale-out network. But long term, we see that the scale-up network also represents a really large growth in TAM, given the fact that we expect the volumes for scale-up to be potentially in order of magnitude larger than the scale-up connections.

So I think on several fronts, we can make the argument that we're still in the early stages. And I don't think there's any doubt in the market about if you can use copper, you will use copper given the advantages for reliability, power and cost.”

This was also stated in terms of how the market will only grow from here for Credo:

“Yes, I wouldn't say it's across the board. I would say that the first step typically is intra-rack, so 3-meter or less connections within the same rack. And this is just recent over the last 6 to 9 months that we've seen traction as our customers start to realize the opportunity to deliver much better cluster reliability and also secondarily better power. And so I would say that we're at the early stages still of having the market expand into rack-to-rack types of solutions. But I do think there's going to be an acceleration in the way that our customers view and use AECs.”

Increasing TAM from Tighter GPU Clusters, PCIe Retimers, LROs and More:

In the opening remarks, the CEO emphasized that Credo is expanding its total addressable market in a few key areas. That message is especially relevant for investors in a hypergrowth stock like this one, where analyst models currently forecast a sharp slowdown — from 274% YoY growth to just 26% over the next three quarters. Analysts have consistently set the bar too low, as shown by Credo’s roughly 30% beats for three straight quarters. Still, the key question remains whether the company can continue sustaining this pace of growth.

Management pointed toward the following:

Packing more GPUs into clusters is a catalyst for AECs:“We also see the trend towards GPU and cluster densification to continue to be a catalyst for an expanding AEC TAM. Over the past year, we've seen customer interest for AECs expand from intra-rack solutions to rack-to-rack solutions.”
Management also pointed toward scale-up as a significant growth driver, which makes sense and is what our thesis is formed on, yet it’s good to see the CEO emphasize this: “And so I think that for us, we'd just like to see the market go faster sooner because the scale-up opportunity represents a significant increase in TAM really over the next two to five years.”
Optical DSPs and LROs: Credo foresees expanding their TAM beyond copper with the goal of doubling optical revenue in FY26. Management hinted they plan to release more products for optical networking at the system-level and 800G LROs. There was also mention of improving the connection between GPUs and memory as a greenfield they plan to go after.
Ethernet Retimers and PCIe Retimers: Credo has recently expanded into PCIe solutions for AI networking which they stated significantly broadens TAM ahead of the shift to 200G per lane scale-up architectures. These products are called Toucan and Magpie.

During the Q&A, the CEO stated the PCIe scale-up opportunity was a bigger opportunity than Ethernet scale-up: “Yes. I would say that the near-term opportunity for us to scale up is really with the PCIe protocol as we see the market moving from PCIe Gen 5 to Gen 6. We do see that AECs will represent a really nice opportunity, both for intra-rack as well as rack-to-rack as scale-up goes row scale”

Conclusion:

We had stated in our Nvidia earnings write-up that the QoQ growth in Nvidia's networking segment should spell good things for I/O Fund members who hold Credo and Astera Labs. So far, so good in terms of the read-through.

What is unique about Credo is not only the hypergrowth that flows effortlessly to the bottom line, but that by my estimation, we are still very early to this trend. I recently said in an interview that networking is what defines the current generation of GPUs, and certainly Credo’s report supports this takeaway.

Typically, I’d be concerned a company like Credo is hitting peak growth, and there could be lumpy quarters; however, management spent a good deal of time on the earnings call going over why Credo is doing so well and why Credo will continue to do well from both a product differentiation standpoint (AECs are in high demand over traditional copper and optical solutions) but also how they plan to expand to meet the fluid needs of intra-rack and rack-to-rack architectures.

This marks the final week of a busy earnings season for the I/O Fund, and Credo saved the best for last. The company continues to hold a prime spot at the top of our aggressive buy list.

Recommended Reading:

Lumentum: Stronger QoQ Growth than Coherent, New $600M Quarterly Revenue Target

Posted on September 3, 2025June 30, 2026 by io-fund

Lumentum has been on our radar for more than one year, as the company supplies components for datacom transceivers and optical interconnects with tech that has caught the attention of heavyweight Nvidia. We’ve been closely monitoring Lumentum and waiting patiently for their EML lasers for 200G to ship, enabling 800G and 1.6T bandwidths.

As discussed in the past, optical interconnects help data centers accelerate data throughput between data centers and inside the data center between servers or racks, while reducing latency and power consumption. AI is driving cloud demand higher from the hyperscalers, leading to more data being created and processed, thus helping drive a need for these interconnects to meet demand for high-speed, low power data transmission in data centers.

Specifically, Q4’s report provided confirmation of the EML laser ramp, as EMLs achieved an all-time high for shipments with revenue more than doubling versus its June 2024 baseline. Management also cited a “substantial” 200G EML order to be fulfilled in the December quarter, although offered little additional clarity on the size of the order.

Overall, Lumentum was quite confident in its growth opportunities from EML lasers, optical circuit switches and co-packaged optics heading into 2026. The company provided a new $600 million quarterly revenue target it expects to reach by next June or earlier, up from $424 million in the current quarter.

Lumentum also was arguably much stronger than peer Coherent with accelerating QoQ growth of 16.1% growth in its data center segment compared to Coherent at 5% QoQ. We break all of this down below.

Revenue Growth Re-accelerating after Tough FY24

Lumentum reported Q4 revenue of $480.7 million, beating analyst estimates by a modest 2.29%. This was a notable uptick on the top-line, with growth of 13.1% QoQ, accelerating 7 points, and 55.9% YoY, accelerating 42 points. This growth begins to support the thesis that Lumentum is past the cyclical low it experienced in FY24, including US imposed trade restrictions that resulted in lost revenue.

Guidance for Q1 FY26 came in at $510 to $540 million, representing 55.8% growth YoY and a slight moderation to 6% QoQ growth. Management attributes the strong forward revenue guide to surging AI workloads and the “shift toward high-speed photonics for hyperscale cloud operators.” Looking beyond the Q1 guide, analysts' expectations for Q2 and Q3 of FY26 reflect growth decelerating to 39.2% and 38.2% YoY, respectively. Regardless, this is still very strong growth.

FY25 revenue came in at $1.65 billion, with growth rebounding to 21% YoY from FY24’s (23%) YoY decline. FY26 is expected to see revenue growth accelerate nearly 20 points to almost 40% YoY, with current estimates at $2.29 billion.

The turnaround is from AI helping to drive a rebound in the Cloud and Networking segment, and is also due to the end of a deep trough in the telecom-exposed segment of Industrial Tech.

Key Segments: Cloud & Networking QoQ Growth Accelerates to 16%

Lumentum has two key segments: Cloud and Networking & Industrial Tech. Cloud and Networking continues to lead as the company’s primary growth source, representing ~85% of total revenue last quarter.

Cloud & Networking Q4 revenue came in at $424.1 million, representing 66.5% YoY growth. Additionally, the segment’s QoQ growth of 16.1% accelerated sharply from 7.6% QoQ in Q3, coming in much stronger than Coherent, where growth decelerated from 10% QoQ to 5%.

Management cited a few factors behind the outperformance in Q4: strong hyperscaler demand driving more than 50% QoQ growth in cloud modules, all-time high in EML shipments, and strong transceiver demand. Of these, EML drove the results this current quarter: “In Components, we achieved an all-time high in EML shipments, nearly doubling the revenue compared to our June quarter 2024 baseline.”

For Q1, management expects Cloud & Networking to be up QoQ, and based on guidance, this would likely correspond to approximately 10% QoQ growth at midpoint. Management indicated there was potential for strong growth starting in the upcoming December quarter: “Recently, we received a substantial order for 200-gig lane speed EML chips, which we expect to fulfill in the December quarter. Overall, we expect 2026 to be a breakout year for laser chip sales of both 100-gig and 200-gig lane speeds.”

Industrial Tech revenue came in at $56.6 million in Q4, representing 6% growth YoY but a (6%) QoQ decline. Management acknowledged that the segment revenue is seasonal and therefore volatile.

For Q1, management guided for Q1 revenue to be approximately flat QoQ, or a marginal 2-3% YoY increase.

It is worth noting that, like many of the AI names we cover, customer concentration is a risk present in Lumentum: two customers currently represent 31% of total revenue. This figure is consistent with the customer concentration reported last year. As the company continues to grow revenues, we would look for this ratio/concentration to decrease, signaling less reliance on any one customer and increased market penetration.

When asked about customer concentration for specific products, such as cloud modules and OCS, they stated that due to their being capacity constrained, it was unlikely they would take on new customers.

Operating Margin Recovery

The re-acceleration in revenues mentioned above drove solid in GAAP gross margin to 33.3%, higher than the 28.8% percentage seen in Q3 and double the 16.6% margin from Q4 FY24. Continued improvement in margins will confirm that management is not only capturing market share in a cost-effective manner but also effectively integrating previous acquisitions in a shareholder-friendly manner.

Non-GAAP gross reiterates this story, as Q4’s 37.8% margin continued to expand versus the prior quarter comp of 35.2% and prior year comp of 27.8%.

It is worth noting that in 2024, gross margins were negatively impacted due to US-imposed trade restrictions which limited the Company’s ability to sell to certain products to one customer. This resulted in roughly $20 million of inventory obsolescence write-offs, a temporary but negative impact to gross margins.

The revenue growth and improved unit economics lead to continued progress on operating margins as well. Q4’s GAAP operating loss of ($8.4 million) represents a (1.7%) operating margin, compared to (8.9%) in Q3 FY25 and (43.3%) in Q4 FY24.

Non-GAAP operating margin of 15.0% expanded nicely compared to prior quarter of 10.8% and prior year quarter of (5.1%). Most of the profit margin is driven by the Cloud and Networking segment, which boasted a margin of 23.6% in Q4, up more than 13 points YoY, whereas Industrial Tech’s margin improved 6 points to just 6%.

For Q1, management guided for continued expansion in adjusted operating margin to 16.0-17.5%, demonstrating continued operational efficiencies alongside strong top-line growth.

Regarding margin expansion, it’s likely to continue the closer Lumentum gets to the $600 million goal with a gross margin of close to 40%: “So because of that and the focus really being on the components business, getting us to $600 million, we should see a nudge-up again on gross margins, getting us close to the 40% that we outlined in the April OFC presentation where we said that we thought at $600 million would be between 37% and 40%. And just given the mix that we're seeing in front of us, we should be at the higher end of that range when we get to $600 million.” The opening remarks reaffirmed this goal: “Cloud revenue is growing well over 20% annually […] gross margins are set to surpass 40%” with the goal of seeing another 220 bps minimum by June 2026.

Adjusted Net Income Margin Continues Path of Expansion

Lumentum reported adjusted EPS of $0.88 in Q4, beating estimates of $0.81 and improving against the $0.57 reported in Q3 and $0.06 reported in the year ago quarter. Q4’s adjusted net income margin of 13.1% reflects continued operational improvements and the fourth consecutive quarter of sequential improvement (from 9.6% in Q3, 7.5% in Q2, and 3.6% in Q1).

As seen below, revenue growth when combined with operating margin expansion leads to sizable increases in the bottom-line returns. Management guided for $0.95 to $1.10 in adjusted EPS in Q1, up nearly 470% YoY, while Q2 is expected to see 176% growth to $1.16.

Cash and Balance Sheet: Positive Operating Cash Flow = Operational Flexibility

As we look at the business and its cash flow, the turn to positive operating cash flow this quarter and continuation of that trend will help alleviate any investors’ concerns about liquidity. Cash flows have been lumpy, though Q4 saw Lumentum report the highest operating cash flow margin in the past two years. However, free cash flows are pressured as Lumentum reinvests to expand manufacturing capacity.

Operating cash flow increased 80% YoY to $64 million, representing a 13.3% margin. This expanded nearly 2 points from 11.5% in the year ago quarter and was a notable uptick from (0.4%) in Q3.
For FY25, operating cash flow was up ~5x to $126.4 million, for a 7.7% margin, improving from a 1.8% margin in FY24.
Free cash flow declined (7%) YoY to $10.1 million, for a 2.1% margin, down from 3.5% in the year ago quarter.
As a result of elevated capex, FY25 free cash flow was ($104.7 million), for a (6.3%) margin, improving only slightly from (8.3%) in FY24.
Cash and equivalents of $877.1 million in Q4, largely in line with the $866.7 million reported in Q3. Debt remained largely consistent with the prior few quarters at $2.57 billion.
Inventory came in at $470.1 million, which continues to grow compared to the $422.9 million reported in Q3’25 and $398.4 million reported in Q4’24. Management stated that the inventory increase will support expected growth in Cloud & Networking. EML laser inventory was quoted as “very low” with management saying they are “basically shipping everything that we can make.”

The company uses significant financing in the form of convertible debt to fund day-to-day operations and investment. This capital structure (e.g. high debt to equity ratio) will amplify shareholder returns but can also put a strain on operating cash flow should the company run into short-term cash constraints.

Another key point to continually monitor will be levels around inventory. Significant increases in inventory levels or increases in DIO (Days Inventory on Hand) may be a warning signal of future write downs with P&L impact. For now, we would give the company a healthy balance sheet rating while acknowledging that Lumentum may need to access capital through markets should liquidity become tight.

Earnings Q&A:

New Medium-term Revenue Target at $600M

Given that Q1 would satisfy Lumentum’s May 2024 guidance to reach $500 million in quarterly revenue in calendar 2025, management has provided a new near-term revenue target, now projecting $600 million in quarterly revenue by Q4 FY26 or earlier.

The new target received a fair amount of attention on the earnings call Q&A, namely regarding what products drive this growth and the outlook for margins. Given the strong language from management, the current takeaway is this guidance could be conservative. Of course, we need a few more earnings reports to see if a higher number materializes, yet strong growth all around is being forecast.

There are four areas that will help Lumentum meet and potentially exceed this forecast. Given the company is expected to report strong 30%+ growth while expanding its margins over the next few quarters, it makes sense to break out management commentary by each product to help organize the many moving parts:

EML Lasers: Management stated there was a large order that is ramping soon: “Recently, we received a substantial order for 200-gig lane speed EML chips, which we expect to fulfill in the December quarter. Overall, we expect 2026 to be a breakout year for laser chip sales of both 100-gig and 200-gig lane speeds.”
This space is highly competitive. When asked why Lumentum is gaining market share over competitors, management stated the following: “Our customers typically report a significantly higher yield on their cloud modules using our EMLs over competitors. That allows us some pricing latitude, which has been super favorable.”
Cloud modules drove half of the sequential revenue growth in the period with over 50% QoQ growth in the quarter. Regarding future growth, it was stated the September quarter would not be as strong, yet would ramp quite quickly in the December quarter and beyond: “The cloud modules will definitely be a step-up. I think we had a really big step-up this quarter, a 50% sequential gain in terms of top line. I think we'll hit a little bit of an ebb here in the next quarter, but then we'll see a pretty dramatic acceleration in cloud modules in our December quarter, March quarter and June quarter.”
Regarding optical circuit switches, it was stated during the Q&A that the ramp will result in significant revenues: “I think that we'll start to see more meaningful revenue, meaning very, very significant revenues in Q1, Q2 and then certainly in the back half of calendar 2026. So it is a ramp. There's some gradualness to it. There's a couple of inflection points. The first inflection point is probably early in '26, but then a more meaningful inflection point in the back half of '26. Right now, we're honestly limited by how many we can build, right? We're trying to ramp this thing very quickly. We see a tremendous level of demand, but we are limited by how much we can supply.” Something similar was later echoed, indicating FY2026 will see more OCS revenue as the year progresses: “We start to see revenue — meaningful revenue contributions in the first half, significant revenue contributions for OCS in the second half.”
Co-packaged optics (CPOs) are not contributing to revenue now yet could materialize into one of the biggest opportunities among all of the components and subsystems that Lumentum supplies. This is one to keep a close eye on.

In the current quarter, management announced “Our commitment to co-packaged optics or CPO is stronger than ever. We just received the largest single purchase commitment in company history […] Our investments in this facility will position us for a significant revenue ramp in CPO by the second half of calendar 2026.”

An analyst asserted in the call that Lumentum has the leading technology for co-packaged optics, which is an important statement given Nvidia is rumored to be rolling out their own internal CPOs in the coming year.

Here’s the original announcement from Nvidia. We covered it here.

“Simon Matthew Leopold

And then as a follow-up, I wanted to check in on the CPO opportunity as well. So that sounds like it's progressing. Last we spoke, it sounded like you were the only approved supplier in the ecosystem for the high-powered laser. Wondering how you're thinking about your position in that market in terms of quantifying as well as your ability to remain a sole source supplier? How is the competitive landscape?”

Optical Circuit Switching Ahead of Schedule and Margin Accretive

We pointed out in Coherent’s post-earnings analysis, Coherent Q4: Data Center Growth Slowing QoQ; Competitive Concerns, that optical circuit switching would increase COHR’s TAM by ~$2 billion, while booking its first revenue last quarter.

Similar to Coherent, Lumentum also booked its first OCS revenue in the fourth quarter with shipments to two hyperscaler customers, yet there were more positives revealed.

What’s more notable is this OCS revenue is two quarters ahead of expectations with more customers; as Raymond James analyst Simon Leopold pointed out: “you previously suggested you'd see first revenue in the December quarter. So this is 2 quarters earlier than what we were thinking and we're thinking one customer, not two.”

Lumentum stated that its OCS order book is expanding with both customers, and it now has a third hyperscaler committed to deploy its OCS in calendar 2026. Management believes that their “leadership in optical performance, particularly in 300×300 form factors has allowed us to capture volume opportunities earlier than competitors.” This hints at possibly winning customers from Coherent, who only said that customer engagement in OCS was “very strong” but did not explicitly state hyperscaler wins.

CEO Michael Hurston also offered clarity on the ramp trajectory for OCS and when revenue is expected to inflect:

“I think the current quarter, next quarter and the December quarter are still ramping. We're ramping because we're building our capacity in Thailand to support the customers. I think that we'll start to see more meaningful revenue, meaning very, very significant revenues in Q1, Q2 and then certainly in the back half of calendar 2026. So it is a ramp. There's some gradualness to it. There's a couple of inflection points. The first inflection point is probably early in '26, but then a more meaningful inflection point in the back half of '26. … We see a tremendous level of demand, but we are limited by how much we can supply.”… We see a tremendous level of demand, but we are limited by how much we can supply.”

Supply constraints aside, OCS is an attractive growth lever for Lumentum as it is accretive to margins. Management said that OCS enjoys margins “significantly above corporate margin averages,” and will be accretive in 1H 2026 as volumes will then begin offsetting dilutive impacts from the factory ramp. This will need to be watched closely considering Lumentum is walking a fine line by simultaneously accelerating in-house OCS manufacturing capacity and expanding indium phosphide production for co-packaged optics (CPO).

800G and 1.6T Shipping, EMLs Sold Out; Company Capacity Constrained

Given the similarities between Coherent and Lumentum as neck-and-neck competitors, it’s interesting to see some of the nuances in commentary for the ramp of 800G and 1.6T over the coming quarters.

Notably, Lumentum shared that it had received a “substantial” 200G EML chip order that it expects to fulfill in the December quarter (Q2 FY26), though management offered little clarity on the size or revenue potential of this order.

Management shared that EML shipments on the 800G transceivers (100G per lane) are continuing to climb, while they are “starting to see the early ramp of 1.6T.” For 1.6T specifically, management said growth would “feather in next year” with the first 200G shipments just now arising. Overall, Lumentum expects 2026 to be a breakout year for laser chip sales on both 100G and 200G speeds.

This is more subtle than Coherent in saying that 800G is “ramping very quickly,” with 1.6T growing on top of that. However, Lumentum made the point of saying that most hyperscalers remain on 800G platforms “probably the next couple of years at a minimum,” as 1.6T is “just getting started,” hinting that the stronger growth for 1.6T may not be seen until 2026 to 2027 and beyond.

Lumentum also dropped an important tidbit relating to competitive pressures, growth and margins. Management said that “customers typically report a significantly higher yield on their cloud modules using our EMLs over competitors,” which has given them “super favorable” pricing. This is likely a key factor behind expanding margins in Cloud & Networking and why capacity is sold out for the year.

Capacity Constraints:

Although Lumentum pointed out that they are the largest supplier in terms of capacity with their Japan factory outputting more EMLs than any other location, capacity constraints were a frequent mention throughout the call.

Lumentum stated outright that while capacity is ramping, demand continues to outpace supply and is expected to remain that way through fiscal 2026. Hurlston said that EMLs are sold out for the balance of the year, and management is being selective in choosing customers based on that limited capacity. He added that completing the transition from 3-inch to 4-inch wafers should help boost capacity, though he did not comment on an estimated completion date. It’s likely that Lumentum will continue to invest in expanding capacity over the next few quarters to help meet high demand, after spending $59 million in capex primary for capacity expansion in Q4, or 12% of revenue.

Limited Cloud Module Customer Engagement

Despite management’s optimism on cloud modules delivering substantial growth into calendar 2026, CEO Michael Hurlston emphasized that three customers would likely be their limit with little room to bring new customers onboard in the near-term. This stems partially from a focus on the highest margin opportunities (OCS being accretive and cloud modules not being accretive), as well as being supply constrained. This is similar to what we discussed in our April analysis, Lumentum at Inflection Point with 20% QoQ Growth in AI-Related Segment, where management pointed out that yield and supply issues were hindering growth.

Hurlston said that Lumentum is focused primarily on these three customers, and that new customer additions in the near term will be minimal. He was straightforward in saying that while Lumentum does expect cloud modules to be beneficial from a revenue growth perspective over the next four to eight quarters, they will drag on gross margins — in the best case scenario, Hurlston said the company expects cloud modules to maybe push 30% gross margins, nearly 10 points lower than corporate gross margins.

Tariff Impacts from Japanese Fabs, China-Sourced Cloud Modules

Tariffs are important to touch upon given the recent fluidity in tariff policy, considering Lumentum is inherently much more exposed to tariffs than Coherent with its supply chain presence in Japan and China.

When asked about the recent 100% tariffs on semiconductor imports and impacts from its Japanese fab presence, management clarified that they “determined that our products are exempted from any of the tariffs that would be applicable in that new guidance” and are “fairly comfortable” that Lumentum will not be impacted.

Additionally, management stated that at the start of fiscal Q4, they believed they would have had “up to a 100 basis point negative impact from tariffs. And yet in Q4, we actually had minimal impact of tariffs.” For Q1’s guidance, management said they included a slight impact “just in case something new pops up in the remaining part of August and September, and we'll see at the end of the quarter where that ends up. But for now, no material changes in our business.”

Lumentum also commented on its expansion efforts in Thailand at its Nava facility, as this is expected to help diversify away from China. CEO Michael Hurlston acknowledged that “a lot” of Lumentum’s cloud modules are sourced from China, and the Thailand footprint should offer some flexibility when it comes to tariffs.

Additional Points:

Apple’s partnership with Coherent: We covered the multi-year partnership between the two in our COHR analysis, with COHR’s revenue expected to see a more meaningful boost in 2H 2026 and into 2027. Analysts asked if Lumentum saw that business area becoming more significant in the future, to which management responded that “Face ID and 3D sensing will be a minimal part of our business on a go- forward basis.”

Conclusion

Lumentum delivered a solid Q4 report with accelerating sequential growth in its Cloud & Networking segment, whereas Coherent fell short of the mark with data center growth decelerating. Q4’s report gave more confirmation and confidence in Lumentum’s EML strength and ramp, with a substantial 200G EML order on the deck for fulfillment in Q2 FY26.

Management appeared quite confident in capitalizing off a trio of growth opportunities from EML lasers, optical circuit switches and co-packaged optics heading into 2026, though there are still lingering concerns about capacity constraints and tariff impacts given manufacturing concentration in Japan and China.

Of the opportunities mentioned above, CPOs are a key catalyst that could provide upside to this stock. CPOs will offer the performance of optical yet with reduced power consumption, with a company like Lumentum supplying the lasers and optical engines that are mounted closer to the switch ASICs. Overall, this would mark a shift in the current AI networking architectures with Lumentum downwind of that shift come 2026. We will be watching this very closely and have a trade setup in mind for this stock.

Damien Robbins, Equity Analyst at I/O Fund contributed to this analysis.

Recommended Reading:

Lumentum: Stronger QoQ Growth than Coherent, New $600M Quarterly Revenue Target

Posted on September 3, 2025June 30, 2026 by io-fund

Lumentum also was arguably much stronger than peer Coherent with accelerating QoQ growth of 16.1% growth in its data center segment compared to Coherent at 5% QoQ. We break all of this down below.

Revenue Growth Re-accelerating after Tough FY24

The turnaround is from AI helping to drive a rebound in the Cloud and Networking segment, and is also due to the end of a deep trough in the telecom-exposed segment of Industrial Tech.

Key Segments: Cloud & Networking QoQ Growth Accelerates to 16%

Industrial Tech revenue came in at $56.6 million in Q4, representing 6% growth YoY but a (6%) QoQ decline. Management acknowledged that the segment revenue is seasonal and therefore volatile.

For Q1, management guided for Q1 revenue to be approximately flat QoQ, or a marginal 2-3% YoY increase.

Operating Margin Recovery

Non-GAAP gross reiterates this story, as Q4’s 37.8% margin continued to expand versus the prior quarter comp of 35.2% and prior year comp of 27.8%.

For Q1, management guided for continued expansion in adjusted operating margin to 16.0-17.5%, demonstrating continued operational efficiencies alongside strong top-line growth.

Adjusted Net Income Margin Continues Path of Expansion

Cash and Balance Sheet: Positive Operating Cash Flow = Operational Flexibility

Operating cash flow increased 80% YoY to $64 million, representing a 13.3% margin. This expanded nearly 2 points from 11.5% in the year ago quarter and was a notable uptick from (0.4%) in Q3.
For FY25, operating cash flow was up ~5x to $126.4 million, for a 7.7% margin, improving from a 1.8% margin in FY24.
Free cash flow declined (7%) YoY to $10.1 million, for a 2.1% margin, down from 3.5% in the year ago quarter.
As a result of elevated capex, FY25 free cash flow was ($104.7 million), for a (6.3%) margin, improving only slightly from (8.3%) in FY24.
Cash and equivalents of $877.1 million in Q4, largely in line with the $866.7 million reported in Q3. Debt remained largely consistent with the prior few quarters at $2.57 billion.
Inventory came in at $470.1 million, which continues to grow compared to the $422.9 million reported in Q3’25 and $398.4 million reported in Q4’24. Management stated that the inventory increase will support expected growth in Cloud & Networking. EML laser inventory was quoted as “very low” with management saying they are “basically shipping everything that we can make.”

Earnings Q&A:

New Medium-term Revenue Target at $600M

EML Lasers: Management stated there was a large order that is ramping soon: “Recently, we received a substantial order for 200-gig lane speed EML chips, which we expect to fulfill in the December quarter. Overall, we expect 2026 to be a breakout year for laser chip sales of both 100-gig and 200-gig lane speeds.”
This space is highly competitive. When asked why Lumentum is gaining market share over competitors, management stated the following: “Our customers typically report a significantly higher yield on their cloud modules using our EMLs over competitors. That allows us some pricing latitude, which has been super favorable.”
Cloud modules drove half of the sequential revenue growth in the period with over 50% QoQ growth in the quarter. Regarding future growth, it was stated the September quarter would not be as strong, yet would ramp quite quickly in the December quarter and beyond: “The cloud modules will definitely be a step-up. I think we had a really big step-up this quarter, a 50% sequential gain in terms of top line. I think we'll hit a little bit of an ebb here in the next quarter, but then we'll see a pretty dramatic acceleration in cloud modules in our December quarter, March quarter and June quarter.”
Regarding optical circuit switches, it was stated during the Q&A that the ramp will result in significant revenues: “I think that we'll start to see more meaningful revenue, meaning very, very significant revenues in Q1, Q2 and then certainly in the back half of calendar 2026. So it is a ramp. There's some gradualness to it. There's a couple of inflection points. The first inflection point is probably early in '26, but then a more meaningful inflection point in the back half of '26. Right now, we're honestly limited by how many we can build, right? We're trying to ramp this thing very quickly. We see a tremendous level of demand, but we are limited by how much we can supply.” Something similar was later echoed, indicating FY2026 will see more OCS revenue as the year progresses: “We start to see revenue — meaningful revenue contributions in the first half, significant revenue contributions for OCS in the second half.”
Co-packaged optics (CPOs) are not contributing to revenue now yet could materialize into one of the biggest opportunities among all of the components and subsystems that Lumentum supplies. This is one to keep a close eye on.

Here’s the original announcement from Nvidia. We covered it here.

“Simon Matthew Leopold

Optical Circuit Switching Ahead of Schedule and Margin Accretive

Similar to Coherent, Lumentum also booked its first OCS revenue in the fourth quarter with shipments to two hyperscaler customers, yet there were more positives revealed.

CEO Michael Hurston also offered clarity on the ramp trajectory for OCS and when revenue is expected to inflect:

800G and 1.6T Shipping, EMLs Sold Out; Company Capacity Constrained

Given the similarities between Coherent and Lumentum as neck-and-neck competitors, it’s interesting to see some of the nuances in commentary for the ramp of 800G and 1.6T over the coming quarters.

Capacity Constraints:

Limited Cloud Module Customer Engagement

Tariff Impacts from Japanese Fabs, China-Sourced Cloud Modules

Additional Points:

Apple’s partnership with Coherent: We covered the multi-year partnership between the two in our COHR analysis, with COHR’s revenue expected to see a more meaningful boost in 2H 2026 and into 2027. Analysts asked if Lumentum saw that business area becoming more significant in the future, to which management responded that “Face ID and 3D sensing will be a minimal part of our business on a go- forward basis.”

Conclusion

Damien Robbins, Equity Analyst at I/O Fund contributed to this analysis.