CoreWeave’s Q1 was a bit of a mixed bag, with the company beating revenue estimates by 5.5% yet guiding Q2 revenue roughly (6.5%) below consensus. Despite the soft guide, management maintained its full year guide at $12 to $13 billion, suggesting a strong ramp into the back half of the year, with revenue forecast to accelerate more than 80 points from Q2’s guided 108% to 190% by Q4.  

Margins contracted across the board in Q1, with adjusted operating margin coming down 16 points YoY to just 1%. Similar to revenue, management remained optimistic on the margin front, projecting 2H operating income growth of >11X versus 1H, and a return to low double-digit adjusted operating margins by Q4. 

Looking through 2026 and 2027, CoreWeave is targeting a rather aggressive ramp in active power, noting that it expects to have a “substantial majority” of its 3.5GW contracted power pipeline become active by the end of next year. For context, CoreWeave currently has >1GW active, adding roughly 150MW in the quarter. Bringing new capacity online is imperative for CoreWeave, as despite its backlog rising 49% QoQ to almost $100 billion, current backlog conversion looks to be largely priced in.  

The challenge remains capex and the strain it is placing on CoreWeave’s balance sheet. Part of the reason that CoreWeave saw weak price action after the report is they raised fiscal year capex; “For the full year, we now expect CapEx of $31 billion to $35 billion. The increase on the low end from our previous guidance is related to increases in component pricing.” 

To simplify this, CoreWeave needs a euphoric market environment for the stock to do well (and likely one where interest rates are going down), as for every $1 made about $2 is spent in capex. The concern is when does the bleeding stop, or will CoreWeave remain on a never-ending build cycle given the rapid iterations around GPUs and system components.  

The second concern is the supply environment is becoming trickier with the increasing complexity of AI systems. CoreWeave’s management described this best when it was stated: “Like the truth of the matter is the limiting factor isn't just power, it's labor, it's memory, it's storage, it's our ability to bring up infrastructure.” Each of those introduces an element of risk, more so when you are upside down on capex. 

CoreWeave is No Longer Just a GPU Provider 

As we pointed out a year ago in our analysis CoreWeave: AI Infrastructure Built for the Next Decade; Upside Down Business Model:  

“CoreWeave brands itself as the world’s first “AI hyperscaler” as they offer both infrastructure and a software platform for developing large language models and deploying them. Being dubbed an AI infrastructure player means CoreWeave must offer a compelling value proposition to attract business from arguably the largest competitors in the world – AWS, Microsoft Azure and Google Cloud. In its S-1 fling, the company points out it was built for AI workloads as opposed to the legacy cloud infrastructure-as-a-service providers that were primarily optimized for the cloud software era and e-commerce era. CoreWeave also asserts that outdated cloud infrastructure leads to lower utilization rates when you factor in usage. its S-1 fling, the company points out it was built for AI workloads as opposed to the legacy cloud infrastructure-as-a-service providers that were primarily optimized for the cloud software era and e-commerce era. CoreWeave also asserts that outdated cloud infrastructure leads to lower utilization rates when you factor in usage.  

The company also offers proprietary software to help achieve higher total system performance and more favorable uptime relative to competitors.” 

This quarter, CoreWeave highlighted more progress in its full-stack software approach, offering customers a higher degree of flexibility in consumption. CoreWeave introduced Flex Reservation and spot pricing this quarter, which aim to help customers budget for and manage unpredictable bursts of demand, with both of these new offerings “immediately oversubscribed.” CoreWeave also recently launched CoreWeave Interconnect in collaboration with Google Cloud, following its SUNK Anywhere and LOTA Cross-Cloud offerings.  

These three services aim to minimize friction in managing multi-cloud environments, helping organizations run AI workloads anywhere in the cloud. CoreWeave says the three “are already proving to be highly effective at capturing increased wallet share.” 

Outside of GPUs, CoreWeave is also seeing strong momentum in CPUs, networking and storage, with both CPUs and networking similarly expected to exceed $100 million ARR and storage noted to be multiplying quickly. Overall, these still remain just a tiny fraction of CoreWeave’s ARR.  

Hyperscalers Becoming CoreWeave’s Main Customers 

CoreWeave’s main value proposition lies within its ability to offer high GPU utilization rates, early access to next-gen systems such as Vera Rubin, and circumventing the hypervisor layer with bare metal servers. However, infrastructure is likely to weigh on margins (and currently is), which is why CoreWeave is aiming to pivot towards a SaaS-augmented model with Omni.  

CoreWeave says Omni enables them to “deploy and operate our full cloud stack in customers' own data centers with their GPUs,” with early interest said to be strong across cloud, enterprise and sovereign customers. While this opens the door for CoreWeave to build higher-margin, recurring revenue streams at third-party data centers on third-party-owned GPUs, layering in on top of its core GPU rental model, the main challenge with this approach is that its main customers are hyperscalers, who do not need this offering.  

For example, Meta and OpenAI are CoreWeave’s two largest customers, contributing ~65% of revenue in Q1, with contractual obligations worth roughly $57.6 billion in total or well over half of its backlog. CoreWeave also has a unique relationship with Google, selling capacity to Google Cloud which will then resell that to OpenAI.  

Although CoreWeave believes its customer base is beginning to diversify, such as with financial services approaching $10 billion, and ten customers committing to >$1 billion in spending, the overall nature of its business remains highly concentrated in the hyperscalers. Thus, the company’s target customer, one that needs both compute and high-margin software add-ons, is not appearing in its current customer base. Realistically, this presents a major challenge of how CoreWeave will be able to monetize a SaaS offering considering two-thirds of its business lies within customers who likely have no need for it.  

Targeting Aggressive Ramp in Power; Carries Execution Risk 

CoreWeave is targeting an aggressive ramp in active power over the next 18 months, as it aims to convert a “substantial majority” of its 3.5GW contracted power pipeline over to being active and revenue-generating. This is part of CoreWeave’s broader goal of reaching >8GW of active power by 2030, or an 8X increase over the next four years. The main question that this begs is how CoreWeave can sustain the level of capex needed for such an aggressive ramp.  

As of Q1, CoreWeave surpassed 1GW of active power, adding more than 150MW during the quarter, while simultaneously increasing its contracted power to 3.5GW, up more than 400M during the quarter.  

Breaking this down, and assuming that management’s commentary applies only to the 3.5GW figure and not subsequent intra-quarter increases in the contracted power pipeline, we can roughly estimate CoreWeave’s trajectory and potential capex costs associated with this buildout. 

Under that framework, and assuming “substantial majority” equates to ~80% of the 2.5GW of contracted but not yet active power, this would estimate CoreWeave bringing on 2GW of new active power by year-end 2027, taking total active power to 3GW. This would be roughly 3X its current capacity. 

Management stated in Q1 that they “remain firmly on track to reach or exceed our target of more than 1.7GW by the end of 2026,” so the above assumption would represent a sharper acceleration of capacity additions in 2027, requiring around 1.3GW of new additions if the 1.7GW active power target is met. However, some analysts are expecting a much more rapid ramp in power, with Oppenheimer penciling in potential new additions of 1GW by Q3 2026. 

CoreWeave also shed a bit more light over its power strategy: “We plan to continue to expand our contracted power footprint through leases while also accelerating our development of self-build sites, which will provide us with greater operational control and long-term financial upside. We expect our first self-build site to come online later this year.”  

CoreWeave’s Leasing Strategy Reduces Capex Shock, But Adds Counterparty Risk 

There are a couple of puts and takes to this approach, notably that the lease-based approach can offer a faster time to power with lower capex, while also opening the door up to execution risk from its counterparts. For example, it was reported in December that Core Scientific’s facility in Texas faced a two-month delay due to weather, with this spilling over to CoreWeave as a result (per management on Q4’s timing miss: “the delays in powered-shell delivery associated with the data center provider will have an impact on our fourth quarter results.”) 

However, the benefit of this lease-heavy approach is that it offers CoreWeave a higher degree of financial flexibility when it comes to managing its balance sheet, which is in rather rough shape with less than $2.3 billion in cash to nearly $25 billion in debt.   

Connecting CoreWeave’s active power target to its projected capex for 2026 illustrates why it is opting to go the lease route versus prioritizing primarily greenfield development.  

This is evident when viewing this via the lens of its deal with Core Scientific. Under that deal, CoreWeave is paying $1.5 million per MW in capex during development; this would project CoreWeave’s capex obligations to be roughly $525 million for the remaining ~350MW of capacity Core Scientific has yet to deliver. 

Compare this to a greenfield 350MW build with construction costs of ~$11 million per MW. Such a build would cost CoreWeave $3.85 billion, meaning the lease approach saves it roughly 86% on capex. This also allows CoreWeave to funnel a majority of its capex dollars into high-end GPUs – at current estimates for GB300 GPUs to cost $33 million per GW per Morgan Stanley, CoreWeave could outfit a 350MW facility with GB300 racks for $11.55 billion, versus around $15.4 billion from the ground-up, or savings of ~25%. For CoreWeave, every dollar of savings matters in this buildout.  

In total, bringing on 2GW of power by the end of next year, such as what we outlined above, could cost ~$66 billion via primarily leases (where capex goes almost directly to GPUs), or $88 billion including construction for GB300 capacity. With Rubin estimated to cost $41 million per MW, per MS, costs could reach $82 to $104 billion for 2GW.  

Interest Payments Surging to $2B Annualized 

The reason that every dollar matters for CoreWeave is tied to its debt, as the company is forking out nearly $2 billion annualized as of Q1 on interest payments, with this figure only set to rise further in Q2. This is one of the core problems with utilizing debt to fund its buildout – interest payments are rising faster than revenue, and reductions in the weighted cost of debt, stated as an 80 bp reduction year-to-date, are not yet enough to slow this train down.  

For Q1, CoreWeave’s net interest payments reached $536 million, or more than $2 billion annualized, rising more than 38% QoQ. For Q2, CoreWeave guided for interest payments to be $650 to $730 million, or up nearly 29% QoQ, and moving up to almost $2.8 billion annualized with no real signs of slowing. On a YoY basis, interest payments will be up more than 2.5X next quarter.  

When looking at CoreWeave’s projected capex needs for 2026 – not even including 2027, which is likely to be much higher considering the active power goals management has laid out that – alongside its thin balance sheet with debt 11X more than its $2.27 billion in cash, it’s clear that CoreWeave is not close to breaking this cycle.  

With nearly 90% of its debt carrying >6% to 15% interest rates and not maturing until 2029 at the earliest, this cycle has years to continue.  

Backlog Reaches a Record $99.4 Billion, Fully Priced In? 

CoreWeave reported record revenue backlog additions in Q1 as it began signing initial Vera Rubin deals, while contracting out capacity expected to come online in 2027. Notably, CoreWeave is largely sold out of 2026 GPU capacity, providing a high degree of visibility into this year’s revenue and ARR guidance, with potential for upside if capacity can come online faster.  

Backlog reached $99.4 billion in Q1, up 284% YoY and nearly 49% QoQ, accelerating from 20% QoQ in Q4 and driven by substantial growth from both existing and new customers. Management added that the backlog is all tied to contracts either currently online, or expected to come online in 2026 or 2027.  

For additional color on backlog, CoreWeave expects 36% of this backlog to be recognized as revenue over the next 24 months, representing roughly $35.8 billion.  

Looking ahead to Q2, backlog should move substantially higher as CoreWeave closed a deal with Meta worth $21 billion through 2032, a $6 billion deal with Jane Street (who is also backing CoreWeave with a $1 billion equity investment), and a multi-year deal with Anthropic in April. 

Turning briefly to ARR, CoreWeave inched its 2026 ARR target higher, now projecting to exit the year with annualized revenue of $18 to $19 billion, raised from $17 to $19 billion previously. Management opted to maintain its 2027 exiting ARR guide at $30 billion, adding that 75% of that ARR is already contracted and booked.  

The challenge here is that CoreWeave’s backlog conversion and ARR targets roughly line up with consensus estimates, suggesting that growth is currently priced in. Taking Q1’s $2.1 billion in revenue with the entirety of the expected backlog conversion (although a portion will be recognized in 2028) equals out to $37.9 billion, whereas current consensus estimates point to 2026 and 2027 combined revenue of $37.6 billion.  

For ARR, CoreWeave’s exit ARR target of $18.5 billion for 2026 suggests December 2026 revenue of $1.54 billion, which, based on ramp dynamics, would likely project Q4 revenue to be around $4.4 billion. This is below current estimates for $4.56 billion in revenue in Q4. The same applies for 2027’s exit ARR of $30 billion, which projects December 2027 revenue of $2.5 billion, or Q4 2027 revenue around the $7.3 billion range. This again is below current estimates for $7.57 billion.  

The takeaway here is that CoreWeave has to build and expand capacity at a quicker rate, placing much more emphasis on capex and funding needs, given current revenue targets over the next seven quarters already look to be largely priced in. Management stated that they have “already secured sufficient power capacity to deliver on our 2027 [ARR] target and expect to continue to add new capacity and customer commitments,” yet that power still has to be brought online and become revenue-generating. 

Financials 

Revenue Misses Estimates yet Growth to Inflect into 2H 

While CoreWeave reported a solid 5.5% beat to revenue estimates in Q1, its Q2 guidance fell short of the mark, with the neocloud guiding for revenue nearly (6.5%) below consensus at midpoint. 

Q1 revenue increased 111.7% YoY to $2.08 billion, maintaining a similar YoY growth pace as the prior quarter. On a sequential basis, Q1 revenue increased 32.2% QoQ, a sharp acceleration from Q4’s 15.2% QoQ print. Strong pricing trends aided growth in the quarter, as CoreWeave noted that “average pricing for the A100s, H100s and H200s and L40s all increased QoQ,” while near-term capacity remains largely sold out. 

For Q2, CoreWeave guided for revenue to be $2.45 billion to $2.6 billion, roughly (6.5%) below consensus estimates for $2.7 billion at midpoint. This would project a slight deceleration in YoY growth to 108.2%, while QoQ growth would also moderate back to 21.5% QoQ.  

Management was straightforward in saying that they do expect revenue to inflect as they cross from Q2 into Q3, setting the stage for a stronger 2H. Given the full-year guide of $12 to $13 billion, CoreWeave is looking at around $7.9 billion in 2H revenue given the implied 1H revenue of $4.6 billion. Current estimates have 2H plotted out (with a bit of upside at $8 billion) at $3.47 billion in Q3 for 154.2% YoY and 37.2% QoQ growth, and $4.56 billion in Q4 for 189.9% YoY and 31.4% QoQ growth. 

This acceleration into 2H is tied to a few factors, with the first simply being CoreWeave’s active power ramp translating to a larger installed base for it to generate revenue from. It is also supported by strong pricing dynamics with management noting that prices were rising across the board for Ampere, Hopper and Blackwell GPUs, as well as a potentially larger mix of higher-priced Blackwell Ultra instances entering the fray. Pricing strength is extending into 2027 as CoreWeave begins to contract out next year’s capacity, providing more legs for revenue growth to remain strong. 

Margin Questions Remain, though Expectations of Improvement in 2H 

Q1 showed a rather bleak picture for margins, with contraction appearing across the board from gross to net margins. However, management expects Q1 to be the trough for margins for the year, with the largest inflection expected to occur crossing from Q2 to Q3, similar to revenue. Management also reaffirmed that they remained on track for sequential margin expansion through the rest of the year.   

Q1 gross margin was 66%, down seven points YoY and two points QoQ. Management chalked this up to timing of capacity and scale. For the timing point, management explained that during data center fit-out for leased capacity, it incurs lease, power and depreciation expenses for one to two months before revenue generation begins. For scale, management explained “So if you think of it as we're running 50 megawatts, and we add 300 megawatts in a quarter, the impact on gross margin is going to be enormous. On the other hand, when you're running 2,000 megawatts and you add 50 megawatts, it's not going to have as material an impact on your gross margin.” 

Q1 GAAP operating margin was (7%), down four points YoY and one point QoQ. Adjusted operating margin was 1%, in line with management’s guide and contracting 16 points YoY and five points QoQ. For Q2, adjusted operating margin was guided to be roughly 2.4% at midpoint, a slight sequential expansion yet still down more than 13 points YoY – more on this and 2H dynamics below. The difference between the two margin figures boils down to SBC. 

Q1 GAAP net margin was (36%), down four points YoY and seven points QoQ. Adjusted net margin was (28%), down 13 points YoY and 10 points QoQ. 

Double-Clicking on Margin Dynamics 

Q1’s call featured some discussion about CoreWeave’s margin trajectory given the Q1 contraction, signs of inflection in Q2 and management’s expectation to return to low double-digit adjusted operating margin by Q4. This would represent more than 10 points of expansion relative to Q1.  

To put this in dollar form, CoreWeave delivered $21 million in adjusted operating income in Q1, guided for $60 million at midpoint in Q2, and maintained its FY26 guide for $1 billion at midpoint. This suggests 2H adjusted operating income of $919 million at midpoint, or >11X what CoreWeave delivered in 1H. Rightfully so, analysts questioned about this sharp inflection and why management has conviction in achieving this, with CFO Nitin Agarwal stating that it comes down to executing its plan of ramping active power, and seeing adjusted operating income begin to outpace revenue growth in 2H. 

On the point of operating margins, CoreWeave downplayed the impact of rising component costs, such as memory. While component price inflation was a factor in the slight FY26 capex raise from $32.5 billion to $33 billion, management emphasized that deal pricing means any such costs are effectively passed on through to customers: 

“We build our contracts to incorporate the cost of all of the components that are necessary to deliver infrastructure. And so by and large, we are insulated from the price inflation on some of the components because we include that in our pricing that we ultimately bring to clients in order to target the margins that Nitin spoke to, right, is up in the mid-20s is how we think about it on a unit basis. … 

And so we've done a really good job of understanding what the components and electricity costs are going to be, we have it structured so that it is effectively passed through when we enter into the contract once again, to ensure that we're able to hit our targeted margins on a unit basis.” 

EPS and Adjusted EBITDA 

Considering the margin contraction this quarter, CoreWeave reported larger losses than expected this quarter. FY26 and FY27 revenue revisions over the past six months illustrate how much farther CoreWeave has moved from profitability, emphasizing that this path remains challenging. 

Q1 GAAP EPS was ($1.40), coming in below the ($1.20) estimate. Q1 adjusted EPS was ($1.12), missing the ($0.91) estimate by more than 22%, and widening from ($0.61) a year ago. 

Q2 is expected to see minimal improvement, with GAAP EPS projected to be ($1.24) and adjusted EPS projected to be ($1.03). Despite the expected progress on the margin front, CoreWeave is estimated to remain unprofitable this year, with Q4 adjusted EPS currently projected to be ($0.34). 

To illustrate how much further CoreWeave has moved from profitability, its FY26 adjusted EPS estimate now sits at ($3.29), down from ($0.23) in November. For FY27, adjusted EPS estimates sit at ($0.63), down from $2.26 in November, with profitability not projected until Q4 FY27. This is likely driven by a handful of factors – gross margin compression from timing of bringing capacity online, possible impacts to gross margin from higher power prices, increased D&A hitting the opex line via technology and infrastructure expenses as CoreWeave expands its GPU fleet, component price inflation, and ballooning interest pyaments weighing on widening operating margins. It’s also important to note that specifically for component pricing, it’s likely that we are currently just seeing a baseline: “Having said that, in the last 6, 9 months, there has been an acute shortage of certain components that have moved up.” 

Turning to adjusted EBITDA, CoreWeave reported $1.16 billion in Q1 for a 56% margin, down six points YoY and one point QoQ. 

Cash Flows, Balance Sheet in Rough Shape due to Elevated Capex Needs  

As expected, CoreWeave’s cash flows and balance sheet are in rather rough shape, though the company highlighted that it has no debt maturities until 2029 aside from self-amortizing contract-backed debt and OEM vendor financing. This gives some leeway for revenue to scale and cash flows to (hopefully) begin improving to help service debt, considering interest payments on said debt are now running at more than $2 billion annualized and expected to reach above $2.5 billion annualized next quarter. 

Q1 operating cash flow was $2.98 billion for a 144% margin, driven primarily by depreciation and changes in accounts receivable. This improved from a 99% margin in Q4 and a 6% margin a year ago. 

Q1 free cash flow was ($4.71 billion) for a (227%) margin, widening from a (159%) margin in Q4 and (137%) a year ago. Capex was $7.7 billion in the quarter, while Q2 capex was guided to be similar at $7 to $9 billion. 

Cash totaled $2.27 billion while debt reached $24.86 billion. Debt will move higher in Q2 as CoreWeave closed a $3.1 billion loan in mid-May as well as $4.5 billion of senior notes in April. Management commented in Q1’s call that they have reduced the cost of debt by 80 bp year to date, though interest expenses are still rising. Overall, CoreWeave has raised >$20 billion year-to-date, including the “first ever investment-grade Delayed Draw Term Loan backed by HPC infrastructure, achieving an A- equivalent rating from Moody's, Fitch, and DBRS.”  

Conclusion 

There are two ways to view CoreWeave, and both deserve a discussion, as the company is at the intersection of the quality AI trade and AI hype. At one point, we had CoreWeave on a Top 15 list, yet as more circular investments unfolded and capex ballooned relative to revenue, our process, which is to seek strong fundamentals, caused us to drop the stock from the list. That circularity is intensifying, as even its non-hyperscaler customer Jane Street is providing a $1 billion loan on a $6 billion sale.  

Key supplier Nvidia also bought another $2 billion in CoreWeave shares in Q1, which creates a triangle where Nvidia’s is the supplier, investor and customer. Right now, the economics are such that 2026 capex is $30-$35B versus $12.5B in revenue at the midpoint, or about $2.60 in capex for every $1 of new revenue. Another risk to consider is CoreWeave’s ARR targets, as 2026 and 2027 targets both suggest potential Q4 revenue this year and next a bit below current consensus estimates, placing the emphasis on accelerating its buildout, and thus capex, to fuel higher growth. 

Overall, macro may matter more to CoreWeave than any of the stock-specific information above. As a heavily debt-funded, long-duration cash-burn story, it will do better in a lower-rate environment. For this stock, if we do enter, it will be 90% technicals.

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. Beth Kindig and the I/O Fund do not own shares in CRWV at the time of writing and may own stocks pictured in the charts.

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

Recommended Reading:

• Core Scientific: Multi-GW Pipeline, New Hyperscaler Interest but Still Tied to CoreWeave
• Dell Fiscal Q1: Agentic AI Creates Tailwind for Traditional Servers and Storage
• Broadcom Offers Strong AI Growth at Scale; Yet Enters Circular Investing
• Nvidia Fiscal Q1: Perfect Quarter, Imperfect Catalysts

Core Scientific represents one of the stronger Miners as they are already earnings revenue on 243 megawatts with another 200 megawatts expected to be earning revenue in the coming months. This helps transition Core Scientific away from being pure speculation as the company is beginning to execute. With that said, Core Scientific must continue to execute to reach its full potential, and this is particularly important because of the company’s debt structure.  

As a reminder, Core Scientific is pivoting from being a Bitcoin miner to offering multisite infrastructure buildouts as a colocation data center provider. Therefore, at the moment, Core Scientific offers a challenging fundamental profile, with operating margins and cash flows still in the red. However, management stated the 243MW currently being billed will result in $350 million annualized colocation GAAP revenue, and raised their expected cash gross profit margin up 500 bps to 82.5% at the midpoint. The takeaway is that Core Scientific can offer enough profitability and visibility to support more growth.  

Notably, Core Scientific has expanded its gross power pipeline to 4.5 GW with 3GW of that pipeline leasable by customers, suggesting annual revenue opportunities of more than $5 billion under current deal economics at full scale. Although this is quite promising, it circles back to execution and debt structure, which is discussed more below.  

Lastly, Core Scientific still remains closely tied to CoreWeave (whose planned $9 billion acquisition of the miner fell through). The first attempt to diversify beyond CoreWeave with a hyperscaler fell through as a deal under exclusivity was allowed to expire. It’s a yellow flag that a hyperscaler deal fell through (raises important questions), although management seems optimistic as they are in discussions with three hyperscaler customers.  

Quick Recap on Miners’ Value Proposition 

Time-to-power is becoming a central bottleneck for AI data centers, as grid constraints rise, connection queues lengthen, putting the emphasis on quick, suitable on-site and behind-the-meter solutions such as Bloom’s fuel cells and GE Vernova’s gas turbines.  

Bitcoin miners offer a third solution for the time-to-power thesis, offering up to several GW of capacity in quick fashion, bypassing interconnection queues for greenfield builds, and offering cheaper electricity costs from long-term power contracts. For example, miners such as IREN and TeraWulf have touted electricity costs around $0.046-$0.047/kWh in the past, compared to PJM’s ~$0.08/kWh in 2025 and commercial sector rates averaging $0.08-$0.22/kWh. For a 400MW data center, electricity expenses at miner rates would be roughly $162 million, versus $280 million to $771 million under commercial sector average rates, or annual savings of ~42% to 79%. 

For neoclouds such as CoreWeave, these lease-based deals and cheaper electricity costs offer a compelling structure to quickly bring significant capacity online to scale revenue, without bearing the construction capex (around $12-14 million per MW) or dealing with power procurement.  

Overall, the value proposition is that miners are cheaper and faster than new, greenfield data center sites that are not energized, yet the downside is that they are capital constrained and may be unable to build-out capacity beyond what is currently in their pipelines. 

According to Core Scientific, they can offer five sites with the first of them ready-for-service (RFS) with a timeline of 18 months or less. Part of Core Scientific’s current strategy is to stop retrofitting the existing infrastructure and to pursue greenfield instead. Thus, the 18 months reflects a (very quick) newer build rather than the unpredictable nature of upgrading older sites. Here is what was stated on the call: “I think the thing that was much more difficult than we certainly gave a credit for was the — was actually executing on brownfield conversions, which is why everything you see that we're doing forward is actually a greenfield site with a very highly standard basis design that allows us to get kind of leverage over our supply chain and be super predictable in terms of our delivery dates.” 

Although this pushed back the timeline on when a Miner like Core Scientific becomes a more viable stock, it also could increase the predictability of deals getting signed and executed as we move further into the 18-month lead time (i.e., 2027). 

Gross Power Pipeline Expanded to 4.5GW, Lots of Execution to Get There 

When we first covered Core Scientific more than a year ago for Advanced members here, Core Scientific: Hypergrowth with 21X AI Segment Growth Potential, the miner’s contract power pipeline spanned 1.3GW, yet today, it is more than 3X higher at 4.5GW. Notably, this excludes the 590MW already contracted by CoreWeave, meaning Core Scientific’s gross power pipeline technically exceeds 5GW, more than double TeraWulf’s 2.3GW and in a similar boat as IREN and Applied Digital around 4.5GW each in North America.  

This 4.5GW power pipeline translates to 3GW of leasable capacity, with 1.5GW of that 3GW figure currently grid-connected. Two sites – Muskogee, Oklahoma and Pecos, Texas – account for the majority of Core Scientific’s pipeline, both with potential to scale to 1.5GW gross power each, or 1GW leasable. 

However, Core Scientific’s current focus remains on delivering its capacity for CoreWeave, ramping from the 243MW billable as of Q1 to 450MW by the end of Q2, with the target of reaching full capacity in early 2027:  

"Across our remaining contracted sites, we will continue delivering billable megawatts over the coming months while scaling execution on the CoreWeave contract, positioning us to deliver more than 450 billable by the end of the summer, while remaining on track to deliver the full 590 megawatts by the early 2027.” 

Executing on CoreWeave’s ramp will be the primary focal point for 2026 and early 2027, as management hinted that delivery of non-CoreWeave capacity within the 4.5GW pipeline will not occur until early 2027: “strategically positioning the business to sign attractive new customer contracts with capacity outside of CoreWeave available for delivery starting in early 2027.”  

What is Core Scientific’s Pipeline Worth? 

We can roughly translate what this power pipeline could suggest for Core Scientific’s revenue potential at 3GW of total leasable power. At roughly $1.4 million per MW per year, or the current run rate of its CoreWeave deal, its entire pipeline (should it materialize and at similar terms) could be worth $4.2 billion in annual revenue potential. Should it advance towards $1.7 million per MW, or in line with Applied Digital’s recent deal, that annual revenue potential moves to $5.1 billion.  

Overall, Core Scientific’s power strategy is all about ‘proactive positioning’, as management put it – securing land, labor and equipment to keep delivery timelines and ready-for-service dates on track for within 18 months, securing gas and behind-the-meter power to enable expansion, and showing prospective customers on-the-ground progress to entice deal-making discussions.  

As you can see, to get to the full pipeline, it comes down to strong execution. 

Shifting to Greenfield Development 

There was one interesting topic of discussion late in Q1’s call about lessons learned from developing multiple sites for CoreWeave, and how that translates into a competitive advantage. On this, management revealed that the advantage lies within their development approach – they are not retrofitting sites, but rather developing them from scratch: 

“I think the thing that was much more difficult than we certainly gave a credit for was the — was actually executing on brownfield conversions, which is why everything you see that we're doing forward is actually a greenfield site with a very highly standard basis design that allows us to get kind of leverage over our supply chain and be super predictable in terms of our delivery dates. 

Brownfield sites are highly unpredictable. They require a lot of customization. It's a lot of effort to try to retrofit an existing building. While sometimes that could be faster, it comes with a lot more complexity.” 

This is a key distinction we raised in our Hyperscaler Power analysis, Why Power is Critical for Data Centers and their Hyperscaler Customers:  

Brownfield sites (retrofitting) are the path other Bitcoin miners are taking as it is cheaper and faster than greenfield, allowing them to convert old Bitcoin mining halls into AI data center capacity at a relatively quick pace. However, greenfield builds – where the developer owns the land, power, and building – offer a higher degree of customization, though at a much higher cost and often with the longest timelines to completion due to permitting, site selection, and grid connection. 

It is that last point where Core Scientific gets its greenfield advantage – it does not have to deal with lengthy site selection or grid connection requestions, with a repeatable development playbook that can pencil in RFS dates within the next 12 to 18 months. It also represents (somewhat of) a faster path-to-market compared to typical greenfield hyperscale builds, which can take >16 months to reach first operations since starting construction.  This is evident in the Muskogee campus with the Polaris deal adding 440MW of contracted energy to its facility, bypassing the grid interconnection queue and opening the door for quick expansion (pending demand and capex). 

However, the main challenge is financing these greenfield builds faster than its miner peers, as these new sites are crucial in diversifying customer exposure outside of CoreWeave. For example, TeraWulf is targeting 2H 2027 delivery of up to 480MW at a single campus in Kentucky, while IREN is aiming to add more than 700MW in 2027. 

$3.3 Billion Financing Helping Accelerate Site Development 

We can roughly infer what Core Scientific’s pipeline would cost to build out, based on management’s estimates for development costs of million per MW. Core Scientific is estimating build costs to be roughly $11 million per leasable MW, meaning that its leasable pipeline of 3GW would cost in the ballpark of $33 billion.  

On this topic, Core Scientific recently closed a $3.3 billion secured senior note raise due in 2031, netting $2.9 billion in gross proceeds, to be allocated across its five sites currently under development.  

Of the proceeds, $2-2.2 billion is expected to help support development of roughly 1 GW in leasable capacity, or two-thirds of its current grid-connected 1.5GW of leasable capacity. It’s important to note that the $2-2.2 billion figure merely represents a 20% cash outlay necessary to land project financing, with this completing the remaining 80%, or ~$8.8 billion.  

All told, the 3GW pipeline would likely require $6-6.6 billion in cash and potentially more than $25 billion in related project financing to be fully developed under a similar structure. This is more than 6X Core Scientific’s current cash balance, meaning debt or project financing will be leaned on quite heavily. The problem with that is high interest – the $3.3 notes carry a 7.75% rate (essentially a junk bond), meaning Core Scientific will add $256 million in interest payments annually, placing further strain on its balance sheet. While it does help avoid diluting shareholders, it opens the door to execution risk, both by Core Scientific and its key customer CoreWeave.  

Importantly, management emphasized that they are not waiting for a deal to be signed before advancing site development, with current capital letting Core Scientific build and target 12 to 14 month ready-for-service timelines. Considering that debt is funding the non-CoreWeave-associated buildouts and that Core Scientific is not waiting for deals to be signed to continue building, analysts questioned what guardrails Core Scientific had on capex:  

“Is there any guardrail on how much CapEx you would start putting forward before getting a lease? 

Adam Sullivan, CEOAdam Sullivan, CEO 

“I mean the way we're thinking about it right now is we want to take the first data hall to full RFS. And as part of that, that means we're securing the labor, securing the trades, we're securing long lead equipment. And we're putting ourselves in a position where if a customer signs really within any time period leading up to the RFS, the first data hall, we can just continue to extend all of that labor that we have secured on site. So that's kind of our guardrail right now in terms of where we sit. But we feel very confident in the strategy and the ability to show the progress that we're making across each of these sites to customers is really what's forcing the engagement here because everyone is incredibly interested in capacity that's getting delivered in '27 right now.” 

The main readthrough here is that Core Scientific does not really have a financial guardrail in place, and instead is banking on a deal being signed, based on high interest in the market and the fact that they are making construction progress. This means that Core Scientific will likely be on the hook for a majority of the construction costs of the new facilities, a shift in strategy from its CoreWeave deal where the neocloud is fronting some of the capex bill for its capacity. Management also noted that they are looking to deploy behind-the-meter power solutions over the same 12 to 14 month timeframe, which, if not included in the above capex portrait, could add $500 to $600 million per GW to project costs. 

The longer that a deal takes to come to fruition, the more capex, and more debt, that Core Scientific will have to incur.  

Touching on Behind-the-Meter Power 

It’s necessary to briefly touch upon Core Scientific’s willingness to turn to behind-the-meter power solutions as a key method of increasing its power pipeline. Based on commentary for the planned Pecos and Muskogee expansions, Core Scientific expects behind-the-meter solutions to account for as much as ~1.86GW across both sites (to reach 1.5GW each), if additional grid power under load study does not pan out.  

One of the main benefits of the miners that we had originally highlighted was that miners already have power secured, yet the main risk here is that Core Scientific’s goal of having two GW-scale sites do not have power secured, and instead may rely on more expensive sources of power.  

Core Scientific noted that behind-the-meter offers a faster time to power than waiting for grid interconnection, which is reasonable considering the scale of these two sites;  however, as noted above, behind-the-meter solutions are not necessarily cheap, and could run as much as $500 million per GW. This could add more than $1 billion to development costs across the two sites, an expensive endeavor assuming power would have to be procured prior to a deal. The other risk is that there is no guarantee that Core Scientific would be able to secure the power necessary for both sites to expand, either via the grid or behind-the-meter. 

Core Scientific’s Crux – No Second Deal (Yet) 

The main drawback, to say, is that Core Scientific has yet to diversify beyond CoreWeave, partially because CoreWeave had attempted to acquire Core Scientific, likely limiting its deal-making ability. Additonally, Core Scientific had a hyperscaler in exclusive discussion across its Pecos and Muskogee campuses, yet the customer's exclusivity expired without a deal being signed.  

Despite that exclusivity agreement expiring, management explained that “three hyperscalers immediately engaged on those same sites, and we are now in active discussions.” CEO Adam Sullivan added that it was “hard to determine the exact reasons why” the original hyperscaler did not follow through with a deal, though Core Scientific believed it was “the best time for us to bring these back to market because hyperscalers were knocking at the door and asking questions about the sites. And we knew we could have an opportunity to bring another hyperscaler into the fray.”  

Given that Core Scientific essentially has taken a step back in the deal-making process, analysts questioned about the three hyperscaler engagements, and if this would be starting the process again from scratch or if there were previous discussions that could accelerate a potential deal. Management confirmed the latter, explaining that it was simply “bringing back both Pecos and Muskogee back to the table. And that's really why we are able to immediately reengage with those customers.” Core Scientifc added that it was also in conversations with AI labs, neoclouds and chipmakers.  

Though there was no indication around when a potential deal could be signed, management believes they are closer to a deal than in Q4 and uniquely positioned to close potential deals. This stems from their focus on having ready-for-service dates within the next 18 months with active construction progress, with five sites expected to have first data center halls ready in 2027:  

How does the negotiation get altered with some of these potential customers when you've secured the supply chain and you're kind of moving forward? Does that accelerate discussions? Does that keep them more engaged?  

Adam Sullivan, CEOAdam Sullivan, CEO 

Yes. I mean it definitely keeps them more engaged. I mean they rarely see sites that come across their desk where there's an RFS time line really within 18 months, but even more so less than that. And so for us, being able to show photos and videos of sites with active construction going on and the list of equipment that are on order that dramatically changes the dynamic of the discussions because this isn't just a photo of a piece of land. This is an active construction site actively progressing towards building a data center.  

Although management did not specifically discuss why the hyperscaler fell through, our readthrough is the change in tone from brownfield to greenfield may be where the delay came in. If you go back about 6 months ago, Miners were attempting to retrofit. According to this earnings call, that is a dead-end of sorts and greenfield is the way forward, which would naturally cause a delay in a deal.  

Financials 

Revenue Inflects in Q1 as Colocation Revenue Ramping  

Core Scientific’s revenue inflected in Q1 as Colocation revenue showed a strong ramp with billable capacity for CoreWeave reaching 243MW, up from 185MW in Q4. Q1 revenue was $115.2 million, up 44.9% YoY and 44.5% QoQ, accelerating from (16%) YoY and (1.7%) QoQ in Q4. 

For a segment breakdown: 

Colocation revenue was $77.5 million, up 804.5% YoY and 147.4% QoQ, driven by incremental capacity delivered to CoreWeave during the quarter. This marked a sharp acceleration from 267.8% YoY and 109.6% QoQ in Q4.  

Within Colocation, lease revenue was $59.2 million, up 892.4% YoY and 136.7% QoQ. Power fees passed through to CoreWeave were $21 million, while maintenance cost ($2.7 million).   

Management added that the 243MW of billable capacity represents roughly $350 million in annualized revenue, with 200MW of incremental billable capacity expected to come online by the end of the summer (Q2). This additional 200MW would represent roughly $288 million in annualized revenue, or $72 million quarterly; however, assuming half lands in Q2 due to the intra-quarter ramp timing, Colocation revenue would roughly estimate to $113 million next quarter, up 45.8% QoQ and 966% YoY. 

Digital Asset (Bitcoin) Mining revenue totaled $37.7 million across self-mining ($30.1 million) and hosted mining ($7.6 million), declining (46.9%) YoY and (22.1%) QoQ. Core Scientific is expecting a “meaningful step down” in miners in 2H as it transitions to Colocation. 

Currently, Q2 revenue is projected to be $134.5 million, accelerating to 71% YoY though QoQ growth would moderate to 16.8%. Q3 is projected to see a further acceleration to 112.8% YoY and 28.3% QoQ to $172.5 million in revenue driven by the capacity ramp. 

For the full year, revenue is currently projected to be $622 million, up 95% YoY, with FY27 estimated to reach $1.04 billion, up 66.5% YoY. Considering Core Scientific is aiming to deliver five sites in 2027 and satisfy the full 590MW for CoreWeave in the early part of the year (representing $850 million in annualized revenue), there is potential for upside to the current revenue estimate if it can contract out some of these sites to new customers. 

Operating Margin Impacted by Impairment Charge, Colocation Margin Dynamics 

Q1 saw gross margin improve double-digits YoY as Colocation takes a larger mix and as Bitcoin operations are wound down. However, impairment charges related to the Bitcoin operations had an outsized impact on operating margin.  

GAAP gross margin was 26.1%, up 15.8 points YoY and roughly flat QoQ.  

GAAP operating margin was (269.4%) due to recording a $266.5 million impairment charge in the quarter, widening from (59.1%) a year ago and (147.3%) in Q4. Excluding the impairment charge, operating margin would’ve been (38.1%).  

GAAP net margin was (301.3%), which was not comparable to 724.6% a year ago or 270.8% in Q4 as both quarters benefitted significantly from changes in fair value of warrants.

It’s also important to touch a bit upon Colocation margins, as power fees passed through to customers are recorded as revenue, yet because they are fully passed through, carry a 0% margin.  

Thus, Colocation reported an 57% gross margin overall in the quarter, up 52 points YoY and 11 points QoQ. However, when stripping out passed-through power costs, Colocation gross margin (lease revenue minus maintenance and other expenses) was 78%, up 21 points QoQ. Management added that they have “increased our target cash gross profit range for the CoreWeave contract to 80% to 85%, up from our original target of 75% to 80%” as they now have “much greater visibility into the associated cost structure given we are now billing for a meaningful portion of the contracted megawatts.” 

EPS 

Driven by the impairment charge, Core Scientific reported a large GAAP loss this quarter, though GAAP profitability is expected as early as Q3.  

GAAP EPS was ($1.06) in Q1, down from $1.25 a year ago and $0.42 in Q4. Adjusted EPS was ($0.11), improving from ($0.13) a year ago and ($0.18) in Q4.  

Looking ahead to Q2, GAAP EPS is projected to be ($0.02), likely accounting for no impairment charges, while adjusted EPS is projected to be ($0.06). Q3 is expected to see GAAP EPS turn thinly positive at $0.01, while adjusted EPS would remain negative at ($0.06); however, this profitability likely assumes no impairment charges, which are a real possibility given the expectation of a significant wind down in Bitcoin operations in 2H. 

Adjusted EBITDA in Q1 was $4.4 million for a 3.8% margin, up from (7.6%) a year ago and (53.5%) in Q4.  

Balance Sheet and Cash Flows 

One of Core Scientific’s advantages in the miner landscape is that CoreWeave is fronting a majority of the capex at up to $750 million ($1.5M/MW), whereas other miners are turning to debt and paying the entire construction/retrofitting costs themselves. However, Core Scientific’s current strategy of progressing greenfield builds pre-contract may require a higher degree of self-funding moving forward.  

Q1 operating cash flow was $249.9 million for a 216.8% margin, driven by the impairment charge and sale of ~$208 million in Bitcoin. This was up from a (56.6%) margin a year ago and 197.3% in Q4. 

Q1 free cash flow was ($136.7 million) for a (118.6%) margin, improving from (162.2%) a year ago and (152.7%) in Q4. Capex was elevated at $389.3 million, or ~3.4X of revenue. 

Deferred revenue was $654.2 million, up from $555.9 million in Q4. 

Cash was $1.0 billion, while debt was $2.1 billion; this does not include the $3.3 billion senior secured notes raised in Q1. Debt and project financing will need to be tracked closely given the current health of Core Scientific’s balance sheet with a ($1.3 billion) deficit. 

Conclusion 

Fundamentally, Core Scientific’s revenue is beginning to inflect as it delivers more capacity for CoreWeave, aiming to deliver an additional 200MW by the end of Q2 to take its total billable capacity to nearly 450MW, more than 75% of the way to its full 590MW obligation. Margins remain negative, though GAAP EPS is expected to potentially shift positive as early as Q3 as Colocation revenue ramps into year-end. 

Core Scientific is making solid progress in expanding its power pipeline, with up to 4.5GW of gross power potential with behind-the-meter and load expansions under study, offering up to 3GW of leasable capacity if fully developed. Pecos and Muskogee are expected to be the company’s primary campuses, both with potential to expand to 1GW of leasable capacity each, and likely the main cornerstones in diversifying exposure outside of CoreWeave to hyperscaler customers. 

While the company is working to progress rapidly with greenfield development of its non-CoreWeave-tied sites, aiming to have five sites ready for service in 2027, capex and debt needs must be watched closely as Core Scientific is funding this development itself. The weak fundamental profile of Core Scientific’s financials makes this stock an Advanced-only momentum play, and one we would only participate in for momentum purposes if we felt it was breaking out. Join Knox this week in his weekly webinar for more information.

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

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. Beth Kindig and the I/O Fund do not own shares in CORZ at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• Dell Fiscal Q1: Agentic AI Creates Tailwind for Traditional Servers and Storage
• Vertiv Q1: AI Infrastructure Story Is Getting More Profitable
• Nvidia Fiscal Q1: Perfect Quarter, Imperfect Catalysts
• Applied Optoelectronics Q1: Management Guides to 141% YoY Growth; Execution Comes Next

MaxLinear is another under-the-radar optical networking beneficiary, supplying a range of components within optical transceivers, with its most notable being optical DSPs for 400G, 800G and soon 1.6T solutions.  

The company is seeing strong demand emerge for its Keystone DSP family, with management raising its 2026 revenue forecast for Keystone by >40% already, from $100-130 million to $150-170 million. Impressively, this growth is being driven by 400G and 800G modules as the 1.6T product, Rushmore, has yet to ramp. 

The pivot upstream to ramping more heavily on 800G and 1.6T is expected to benefit both revenues and margins from higher ASPs, but it also opens the door for a prolonged revenue runway as 1.6T growth is expected to maintain strong through 2027 with attach rates increasing with larger GPU systems.  

Fundamentally, margins have remained pressured and cash flows are quite thin, but there are some green shoots emerging as MaxLinear is forecasting a return to GAAP operating profitability in Q2 for the first time in three years.  

Brief Product Overview 

MaxLinear supplies a range of key components within optical transceivers, AECs and other connectivity solutions, such as PAM4 DSPs, transimpedance amplifiers (TIAs) and storage accelerator SoCs. While AI data centers (Infrastructure) are rapidly becoming MaxLinear’s main growth outlet, it also serves broadband, wireless, automotive and industrial end markets.   

• Keystone Optical DSPs 

MaxLinear’s Keystone family spans twelve PAM4 DSPs, including both 400G to 800G with integrated drivers for EML/SiPho transceivers, bare die options, and driverless DSPs. MaxLinear also has introduced variants of Keystone to target both AOCs and AECs as well.  

The company says the third-gen DSPs offer ‘best-in-class’ power consumption, enabling 7W 400G optical modules and 13W 800G designs. The Keystone DSPs can also be paired with MaxLinear’s Topanga and Washington TIAs to provide complete solutions for customers.  

Keystone was stated this quarter to be ramping at multiple major hyperscale customers across the US and Asia for both 400G and 800G scale-up and scale-out applications. It also underpins MaxLinear’s optical data center momentum this year into 2027, with management seeing DSPs as their outright #1 TAM.  

Keystone is expected to drive the near-term growth, per the opening remarks: “We also expect a step function data center revenue increase beginning in Q2 with expected strong upside as run rates expand into 2027. At the center of this data center momentum is our Keystone PAM4 DSP optical transceiver platform. Keystone is now ramping at multiple major hyperscale customers across both the U.S. and Asia, supporting 400G and 800G — 800G PAM4 deployments for scale-up and scale-out applications.” 

• Rushmore 200G PAM4 DSP 

MaxLinear unveiled Rushmore last year, its 200G/lane PAM4 SerDes and DPS supporting 1.6T optics and active copper cable deployments. Rushmore is compatible with both Ethernet and InfiniBand, and will enable <25W 1.6T optical interconnects with low latency, and optimized performance across all laser technologies from EMLs to SiPho. MaxLinear says that when paired with its Washington TIA, Rushmore will “provide the best performance on the market.” 

MaxLinear added in Q1 that customer engagement for Rushmore is accelerating faster than expected, with production ramps expected in late 2026 with strong revenue growth continuing into 2027.  

Here is what was stated in the opening remarks: “So moving forward to 1.6T, the critical thing to keep in mind is that there is enormous confidence out there. We're shipping Keystone into major data centers today, and they're ramping very strongly in 2026. And we have now rolled out our 1.6 Terabit Rushmore product and Annapurna family for electrical applications.  

And I think that this level of execution a part and the success with the cloud relationships, module partnerships and the call and interop completion is creating a far more pull for 1.6T participation than I would have guessed at this point in time. So in a sense, we hope that by the end of the year, we'll have called on 1.6T and start transitioning […]” 

• Washington 200G TIA 

Transimpedance amplifiers (TIAs) are current-to-voltage converters, amplifying the current generated by photodetectors within optical modules to voltage signals that can be processed by DSPs. For high-speed optics such as 1.6T transceivers, TIAs play an important role in helping maintain signal integrity at faster rates. 

MaxLinear recently unveiled its four lane/200G Washington TIA at the end of April, designed for 1.6T transceivers and capable of supporting fully retimed, half-retimed, and linear interfaces, from LRO/LPO, NPO and CPO applications.  

Washington interoperates with PAM4 DSPs from all major DSP vendors, simplifying integration into existing system architectures, while also complementing MaxLinear’s portfolio of DSPs and drivers to allow customers to build and customize systems for their specific power, performance and reach requirements. 

As discussed below, the ability for the TIA to be a component in any architecture is key: “But honestly, I mean, the TIA is beyond the TIA for Rushmore, right? If you think of an LPO strategy that the TIA is a fundamental block. If you think about LRO strategy, the TA is a fundamental block.” 

• Annapurna scale-up retimer 

Annapurna is MaxLinear’s high-performance 224G PAM4 scale-up retimer, helping boost reliability of copper connectivity at 224G/lane speeds. We recently covered this challenge in our free newsletter, Inside Nvidia’s $4B Optical Strategy—and Why CPO Changes Everything, where effective length of AECs shortens from 100G to 200G speeds due to signal degradation. 

Annapurna is available in both eight and sixteen lane configurations to support both 1.6T and 3.2T AEC and on-board retimer deployments, and compatible with ESUN, UALink, and Ultra Ethernet protocols. Currently, MaxLinear is playing on the Ethernet side with Annapurna but management explained they are laying the groundwork to play wherever the retimer market moves in the future, such as with NVLink Fusion. 

The CEO stated the retimer opportunity is humungous: “So especially, I know there's a lot of hoopla about AECs because of success of one very successful company on AECs. But if you look at the market size, the opportunity for a silicon player, the AEC, the retimer market electrical for AI scale it inside the compute server is humongous. — as the speeds increase.” 

• Panther storage accelerator SoC 

MaxLinear’s Panther V storage accelerator enables ultra-low latency direct memory access between storage accelerators, NVMe SSDs, and network cards, bypassing host memory to reduce memory bandwidth consumption in AI workloads. It also offloads compute-intensive compression, deduplication or other tasks from the host CPU to accelerate performance. With on-chip SRAM, Panther V enables high-speed data movement across the storage tier, with its 450Gb/s throughput offering 2X the performance of its Panther III SoC.  

MaxLinear is actively sampling Panther V with key customers, with management forecasting storage accelerator revenue to at least double in 2026 over 2025 based on current engagement and design win activity across Tier 1 network and cloud providers: “And this is just the beginning of our Panther product — Panther road map product family. So we expect this year the revenues to double. We have said that before. And hopefully, next year as well, we got very strong growth based on the visibility we have.” 

The following context on Panther was also shared during the Q&A regarding its nuances at the product level, in addition to serving the massive memory market: “So the big benefit of Panther is it's an accelerator, so it reduces latency dramatically and the power efficiency that brings to it. So it enables much more capability than just memory compression, right? So I really feel that the performance part related to low latency, high-bandwidth access enablement that Panther provides is the key differentiator.  

Thus far, our use of Panther has been really at the enterprise appliance level, if you will. But now these enterprise storage appliance are getting increasingly deployed into mainstream cloud centers. So I really feel there's much more to come with Panther V and Panther VI in the future.” 

Optical DSP Demand Accelerating, Keystone Revenue Target Raised 40% 

It should be no surprise that optics demand is accelerating, as evidenced by recent earnings reports from Lumentum, Coherent and Applied Optoelectronics, and this accelerating demand is driving rapid growth for MaxLinear.  

While we are just one quarter into 2026, the strength of customer orders and visibility into optical program ramps at customers has already led MaxLinear to increase its optical data center growth forecast by 40%. MaxLinear now projects optical data center revenue between $150 million to $170 million for 2026, raised from its prior forecast from Q4 for $100 million to $130 million. Though management did not provide a 2025 number, analysts implied 2025 optical revenue of $60-70 million, so this could represent up to ~180% YoY growth.  

When questioned about intra-quarter dynamics driving this steep >40% raise (and rightfully so), management explained that they were being conservative with the initial guide and optimistic for stronger growth in the second half.  

“Now with all the visibility and the lead times that are necessary for providing the product, we have very good visibility and the ramps are setting in very nicely, both across 400-gig and 800-gig solutions. So I just think it's all about timing of the ramps and the success of the calls and our ability to scale up to meet the demand that the surging demand we are seeing now. 

The important readthrough here is that MaxLinear’s main optics growth story this year is still being driven by 400G and 800G, and is not yet benefitting from higher ASPs with 1.6T. This suggests that its optics growth runway is likely to extend and expand as both 800G and 1.6T ramp through 2027 and take a larger share of units and revenue, coinciding with Rushmore’s ramp – more on this next. 

Despite being a small fish in a big sea, MaxLinear has a few advantages on its side – it has design wins across all optics module vendors globally, with CEO Kishore Seendripu explaining that the “success of Keystone makes us an incumbent, right? And the power of incumbency is the ability to have the relationships with the cloud customers, the module makers, the confidence in your ability to supply and the quality of your product.”   

Additionally, MaxLinear has optical DSP growth coming from both hyperscaler-owned designs (direct qualification) as well as merchant solutions from module vendors, with CFO Steven Litchfield saying that growth is coming both from “hyperscaler-owned designs [and] through module vendors providing a merchant solution.” However, MaxLinear has not been upfront about which hyperscalers it is directly engaged/qualifying with and shipping to, rather emphasizing that they believe they are “only halfway there to our end data center diversification across all the hyperscalers.” This implies that hyperscale engagements may be more limited, or that there is still room to land additional hyperscalers in the future. 

This positions MaxLinear quite well for growth even as the industry continues to face tight supply constraints, as exposure across the vendor landscape should mean that it is rather insulated if individual suppliers face headwinds to growth from these constraints. Despite having this broad customer exposure, MaxLinear does expect to remain fairly concentrated on a few end customers through its ramp this year, expanding into 2027.  

Touching on 1.6T Dynamics and (Stiff) Competition 

There are a couple puts and takes for MaxLinear’s participation in 1.6T optical modules. On the positives, management hinted that demand is already shaping up much stronger than expected, aligning with commentary from transceiver vendors, with 1.6T boding well for growth via higher ASPs and margins.  

First on demand, CEO Kishore Seendripu explained that MaxLinear’s cloud and module vendor relationships, along with Rushmore’s interoperability with any DSP is “creating far more [of a] pull for 1.6T participation than I would have guessed at this point in time.” This is expected to “have an uplifting effect on our revenues and gross margins even as our market share expands” as mix shifts to 1.6T due to the higher ASPs, along with potential for more unit growth. MaxLinear hammered this point home by emphasizing further that 1.6T “will actually expand our ability to garner more revenues and more market share.” 

However, 1.6T is not appearing in growth (yet) and MaxLinear was straightforward in noting that they are not the first with 1.6T against their two incumbent competitors, which may present a challenge in catching up during the ramp. These two competitors are most likely to be Marvell and Broadcom. Broadcom will likely pose a more substantial competitive threat as its Taurus DSP platform is scaling to 400G/lane for 1.6T support (and upcoming 3.2T modules), double MaxLinear’s Rushmore and Marvell’s Ara platforms at 200G/lane. While competition against these two incumbents will be challenging, considering MaxLinear’s size, even a tiny gain in market share could translate to substantial revenue growth.  

1.6T revenue may begin appearing later this year, but the largest contributions will likely be tied to Rushmore’s ramp geared for 2027. Though MaxLinear has not provided much insight into the degree of ASP uplift from Keystone to Rushmore, a rough assumption for a ~30% uplift and a similarly-sized but accelerated ramp versus Keystone could see Rushmore quickly ramp to >$200 million in revenue by 2028. For comparison, Keystone launched in early 2023, and is scaling to >$150 million in roughly three years.  

Broader industry dynamics suggest that this shift upstream from 400G to 1.6T could be quite lucrative, looking beyond the simple ASP growth story. This is because MaxLinear will soon be playing in a much larger market of growth, and also because optics attach rates are expected to increase rapidly as Nvidia’s Vera Rubin platform comes online. 

Estimates from Goldman Sachs earlier this month suggest that 400G is likely accounting for just 5-9% of the market, while 800G is around the 20% level in Q1 and Q2, meaning MaxLinear’s 2026 optics growth story is being driven by the smaller third of the market. Shifting to 800G later this year and 1.6T into 2027 would see MaxLinear move to participating in a higher-value, larger market with both speeds expected to account for as much as 60% of the market by late next year.  

Source: Goldman Sachs 

Outside of shifting from a (declining) 400G market into mainstream 800G and 1.6T markets, growing attach rates for optical modules with Rubin further support strong growth. With Blackwell and Blackwell Ultra, GS estimates optics attach rates of roughly 1:2 to 1:3 depending on a two or three-layer networking topology, but sees this doubling to 1:4 to 1:6 with the VR200 rack. This combination of doubling attach rate translating to higher unit volumes and higher optics content per rack alongside ASP growth with 1.6T offers a strong tailwind for growth into 2027.   

MaxLinear Plays in Scale-Up, Scale-Out and CPO 

It’s important to touch briefly upon scale-up and scale-out demand, as optical transceiver demand is more heavily weighted towards scale-out applications due to copper’s physical limitations over longer distances at 200G and faster speeds.  

MaxLinear is seeing Keystone ramp for scale-up and scale-out with strong growth across both, and engagement across the two for Annapurna and Rushmore. These two products, targeting electrical retimers and AECs, is likely to be where MaxLinear’s scale-up growth appears as the majority of the optical transceiver TAM (70%) is weighted towards scale-out.  This scale-out focus is key over the medium-term as optical transceiver content could nearly triple from ~$173,000 in Blackwell Ultra to nearly $500,000 in the Rubin Ultra NVL144. 

Moving to CPO — as the industry shifts towards CPO, first revenues among the optics stack are expected to be realized in scale-out applications, though as we noted in our Lumentum analysis for Premium subscribers, scale-up opportunities could be larger. On this note, MaxLinear is preparing for a range of optical outcomes, with its Washington TIAs underpinning a ‘full platform’ approach for CPO or other packaged-optics solutions, such as LPO or NPO:  

“So the CPO market, if [customer are] going to be bare bones, then the TIA and driver is a natural fit. If they go more sophisticated on the DSP-based one, we already have the platform offering. But the real question comes, as you go towards XPO, CPOs and the various manifestations of it. So the full offering is super important. So Washington is the first step in the direction of a fundamental platform that will have multiple derivatives and incarnations.” 

The catch here is that MaxLinear does not expect CPO “to be a huge part of our revenues” over the next couple of years, stating they think they are “3 years out from determining” how the CPO market plays out. Regardless, it will be something to pay close attention to as CPO ramps are on deck for the primary module vendors later this year with further growth expected in 2027. 

Infrastructure Growth up 35% QoQ 

Putting this all together, MaxLinear is seeing robust growth arise in its Infrastructure segment, with Q1 revenue of $62.8 million up 136% YoY and notching one of the strongest QoQ growth rates in the AI industry this quarter at 35% (though arguably at quite a small scale).  

Additionally, data center growth is expected to see a “step function” increase in Q2 with strong upside expected into 2027, and based on commentary, this is likely tied primarily to Keystone. Currently, this is being modeled above at ~31% QoQ, or an increase to 50% revenue share from 46% in Q1. This would also maintain a similar YoY growth rate as Q1 at 138% YoY.  

Considering that Keystone has multiple customers progressing with ramps, more programs expected to layer in later this year and potential initial contributions from Rushmore as well, there is potential for MaxLinear to sustain strong sequential growth through year-end.  

Management hinted that they “absolutely” expect more upside to that $150-170 million optical forecast as programs reach full run rates — assuming that growth does not moderate following Q2’s step-function increase but instead remains robust at ~$20 million QoQ through Q4, this could project Q4 Infrastructure revenue to be roughly $122.5 million, up 163% YoY. This would represent approximately a $500 million annualized run rate, double its current rate at $250 million. 

Financials 

Revenue Growth Accelerating

MaxLinear’s revenue has since recovered from the stiffer headwinds it had faced in 2024, where it had seen (35%) or larger quarterly declines across all four of its segments. Q1 revenue was $137.2 million, up 43% YoY and roughly flat QoQ, marking a slight deceleration from 48% growth in Q4. The flat QoQ growth highlights the strength of Infrastructure and Keystone, as Broadband revenue saw a sharp (24%) QoQ decline.  

For Q2, MaxLinear guided for revenues between $160 to $170 million, representing a reacceleration to 51.7% YoY at midpoint, with QoQ growth similarly accelerating to 20.3%. This is expected to be primarily driven by Infrastructure revenue where MaxLinear projected a “step-function” increase stemming from strong optical interconnect demand, alongside growth from all four segments.  

Looking at the second half of 2026, consensus estimates currently point to growth moderating to the 30% range, exiting the year at 33% growth, a sharp ~18 point deceleration from Q2’s guide. This comes from consensus pointing to sequential dollar growth of just ~$8 million in both Q3 and Q4, a substantial step down from Q2’s guided $28 million at midpoint.  

However, there are multiple signals that suggest MaxLinear could exceed these estimates and maintain strong sequential growth in the back half of the year. Given the demand signals we have been seeing across the optical transceiver landscape, there should be few reasons that MaxLinear cannot maintain rather robust Infrastructure growth as outlined above, aside from a lack of execution with the 400G and 800G ramp with 1.6T on deck.  

Also layering into growth will be large-scale deployments later this year for single-chip fiber PON and WiFi 7 platforms at a second Tier 1 service provider in North America with additional ramps in Europe.  As such, maintaining ~$20-$30 million sequential dollar growth through 2H (which could come from Infrastructure alone) could see MaxLinear exit the year with quarterly revenues above $200 million. 

For the full year, consensus points to growth of 40.5% to $657 million, but the scenario discussed above for ~$25 million QoQ in 2H at the midpoint would place FY26 revenue at $707 million. 

Consensus estimates also point to revenue growth decelerating rather sharply to 20.7% to $791.3 million in 2027, yet expectations for “strong upside as run rates expand into 2027” within the data center and 1.6T ramping both suggest revenue could land significantly higher. This does not include MaxLinear’s first XGS-PON win with a hyperscale data center which management explained could be “quite a bit of needle mover even in the next year itself in the second half on a run rate basis.”  

Key Segments 

MaxLinear reports in four key segments: Infrastructure, now its largest segment as of Q1, Broadband, its historically largest segment, Connectivity, and Industrial/Multi-Market. 

As noted above, Infrastructure revenue was $62.8 million in Q1, accounting for 46% of revenue. This marked a 136% YoY and 35% QoQ increase, a sharp acceleration from 76% YoY and 15% QoQ in Q4. Assuming a step up to 50% revenue share in Q2 as the main growth driver next quarter, Infrastructure revenue would be roughly estimated at $82.5 million, up 138% YoY and 31% QoQ. 

Broadband revenue was $43.6 million, accounting for 32% of revenue. While revenue did increase 6% YoY for the segment, sequential growth was poor at (24%) QoQ. Management said there was a seasonality component to this QoQ decline, but the segment is expected to start growing in Q2 and into 2027, supported by fiber PON ramps. 

Connectivity revenue was $18.6 million, accounting for 14% of revenue. Growth was rather soft, down (8%) YoY but up 3% QoQ. Management sees wireless infrastructure momentum improving due to increased investments in 5G ran, transport overhaul and backhaul to support cloud and edge AI connectivity.  

Industrial and Multi-Market revenue was $12.2 million, accounting for 9% of revenue, and up 47% YoY but down (13%) QoQ. 

Margins Negative but Signs of Improvement in Q2 

MaxLinear’s GAAP margins have been quite heavily pressured down the line, yet Q2 is showing a notable shift as management guided for GAAP operating margin to jump towards positive territory. This is likely driven by the step function data center growth and margin tailwinds carried by higher speed optical components.  

GAAP gross margin was 57.5% in Q1, up from 56.1% a year ago and roughly flat QoQ, while adjusted gross margin was 59.5%, up less than a point YoY and again roughly flat QoQ.  

For Q2, MaxLinear guided for GAAP gross margin of 56-59%, up 1 point YoY and flat QoQ, and adjusted gross margin of 58-61%. Management noted that there are some headwinds to gross margin related to rising wafer costs and packaging, but they “will certainly continue to see nice benefits on the gross margin side as infrastructure gets to be a larger percentage of our business.” 

GAAP operating margin was (12.5%) in Q1, ticking slightly lower from (10.9%) in Q4 but marking a solid improvement from (48.1%) a year ago. Adjusted operating margin was 15.9%, slightly lower from Q4’s 16.2% but up from (1.7%) a year ago.  

GAAP operating margin is where green shoots are arising in Q2, with management guiding for a thin but positive 0.5% margin at midpoint. This would represent a strong 13 point sequential increase and MaxLinear’s first positive GAAP operating margin in over three years. Adjusted operating margin was guided to be 21%, a five point sequential increase and up nearly 14 points YoY. This large delta between GAAP and adjusted operating margins is primarily due to high SBC and some acquisition and integration-related costs. 

GAAP net margin was (32.9%) in Q1, as MaxLinear recorded a rather large $26.5 million income tax provision in the quarter; this compared to (51.8%) a year ago and just (10.9%) in Q4. Adjusted net margin was 14.2%, up from (4.6%) a year ago and 12.7% in Q4. 

GAAP Profitability Expected in 2H  

Given the guide for operating margin to shift back to positive territory in Q2, earnings are expected to soon follow.  

Driven by the income tax provision, GAAP EPS was a wide ($0.52), missing estimates for ($0.21) and only minimally improving from ($0.58) a year ago. Adjusted EPS was $0.22, beating the $0.18 estimate and increasing from ($0.05) a year ago. 

For Q2, GAAP EPS is expected to inflect towards profitability, with consensus pointing to just ($0.05). MaxLinear is expected to see positive GAAP EPS in both Q3 and Q4, though remaining very thin. Adjusted EPS is expected to be $0.33 in Q2, up 1,547% on a small comp of $0.02.  

For the full year, GAAP EPS is expected to be ($0.53), driven by Q1’s loss, while adjusted EPS is expected to be $1.34, up 331% YoY.  

Cash Flows and Balance Sheet 

Operating cash flow dipped to negative territory after three quarters positive, though MaxLinear had a strong reason for this – substantial prepayments for wafers to support increasing demand for data center products with increasing 2H backlogs.   

Q1 operating cash flow was ($8.9 million) for a (6.5%) margin, up from (11.9%) a year ago but down from 7.6% in Q4.  

Q1 free cash flow was ($11.1 million) for an (8.1%) margin, up from (14%) a year ago but down from 4.9% in Q4. 

Cash and equivalents totaled $62.5 million, while debt was $123.8 million. 

Inventories were $85.8 million, roughly flat YoY but up from $78.1 million in Q4. 

Conclusion 

MaxLinear is forecasting strong optical data center revenue growth from 400G and 800G products via Keystone, with its 1.6T focused Rushmore ramping later this year into 2027. Management has already raised its 2026 optical data center revenue forecast by >40% from $115 million to $160 million at the midpoints, with a step function increase expected next quarter. Ramping 1.6T later in 2026 and into 2027 presents further opportunities for growth to remain strong considering the ASP uplift and potential for increased content and attach rates within upcoming rack-scale systems.  

Fundamentally, while MaxLinear stands out for its 35% sequential growth in data center-driven revenue this quarter, it arguably is much weaker down the income statement than other optical beneficiaries such as Lumentum. Margins have been quite weak and Q1 did show a larger GAAP loss. Q2 is expected to right the ship and put the company potentially on a path to GAAP profitability in 2H for the first time in over three years.

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.

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

Recommended Reading:

• Arm: Computex Update, CPU Core Demand Hinted at Being Higher
• CoreWeave: Revenue Inflecting in 2H, Margin and Profit Questions
• Core Scientific: Multi-GW Pipeline, New Hyperscaler Interest but Still Tied to CoreWeave
• Dell Fiscal Q1: Agentic AI Creates Tailwind for Traditional Servers and Storage