Bloom Energy Corporation Class A Aktie

Good day, everyone, and welcome to the Bloom Energy Second Quarter 2026 Earnings Call. Just a reminder that today's call is being recorded.

At this time, I would like to hand things over to Mr. Michael Tierney. Please go ahead.

Thank you, and good afternoon, everybody. Thank you for joining us for Bloom Energy's Second Quarter 2026 Earnings Call. To supplement this conference call, we furnished our second quarter 2026 earnings press release and supplemental financial information with the SEC on Form 8-K and have posted these materials, which we will reference throughout this call to our Investor Relations website.

During this conference call, both in our prepared remarks and in answers to your questions, we may make forward-looking statements that represent our expectations regarding future events and our future financial performance. These include statements about the company's business results, products, markets, customers, strategy, financial position, liquidity and full year outlook for 2026. These statements are predictions based upon our expectations, estimates and assumptions.

However, as these statements deal with future events, they are subject to numerous known and unknown risks and uncertainties as discussed in detail in our documents filed with the SEC, including our most recently filed Forms 10-K and 10-Q. We assume no obligation to revise any forward-looking statements made on today's call.

During this call and in our second quarter 2026 earnings press release and supplemental financial information, we refer to GAAP and non-GAAP financial measures. The non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between the GAAP and non-GAAP financial measures is included in these materials, which are available on our Investor Relations website.

Joining me today on the call are K.R. Sridhar, Founder, Chairman and Chief Executive Officer; and Simon Edwards, our Chief Financial Officer. K.R. will begin with an overview of our progress, and then Simon will review financial highlights for the quarter. After our prepared remarks, we will have time to take your questions.

I will now turn the call over to K.R.

Good afternoon, everyone, and thank you for joining us. It took Bloom 21 years to deliver its first $1 billion year in 2022. It took us another 3 years to double our 2022 revenue. Now we are guiding to double that revenue in just 1 year, having achieved our first $1 billion quarter. Bloom Energy's business is accelerating. We are expanding at breakneck speed and continue to grow more profitable.

We have successfully demonstrated the earning power of our agile business model, doing exactly what we built it to do. And through all of it, we kept the promise that we make to our customers. Bloom will not be your bottleneck. We deliver power at AI speed and enable our customers to grow. This makes us a vital strategic partner and builds customer loyalty. And demand is compounding. New customers are arriving at a faster pace than ever. Landing a mega customer used to take us years. A moat that protected the incumbents. Now the proven success of our technology, coupled with an increasingly urgent need for efficient, clean and reliable power has collapsed the time from first engagement to first order. Because we can book, ship and convert orders to revenue inside the same fiscal year. That demand shows up in results now while also adding to and diversifying our backlog.

Notably, just this year, several customers who had alternative solutions in place, abandoned them and came to Bloom. Once customers are in, they see the total economic value of our solution and place more orders. We have booked major customers as of today who are not in our reported backlog at the end of last year for whom we will ship some of our systems this year. These customers are now facing longer-term orders leading to our backlog growing at a faster pace than revenue.

Let me repeat, leading to our backlog growing at a faster pace than revenue. Bloom Energy has emerged as a standard for on-site power as we predicted we would in our third quarter call last year. At that time, we had just announced our first direct hyperscaler customer, Oracle, and delivered them power for a data center within 55 days. Today, all the major U.S. hyperscalers and over a dozen U.S. neoclouds, AI labs and colocation data center operators have validated and approved our power solutions for their AI factories.

Our commercial and industrial business continues to grow. We are the standard for on-site power for hospitals, factories, telecom providers, university campuses and retail stores, but it took us nearly a decade to become the accepted solution in these verticals. Contrast that to AI data centers where we have become a standard in less than a year, I founded Bloom on the conviction that on-site power would be essential to powering the world and ushering in the digital transformation, and we have built our company to offer the best on-site power solution that removes friction for our customers. We are clean, and reliable and fast and affordable. Customers do not have to choose or compromise. And over time, we are reducing friction and turning headwinds to tailwinds.

Let's take a moment to discuss 4 friction points: capital, community permitting and speed. Start with capital. For a century, the cost of power plants and the grid was spread across millions of ratepayers and amortized over decades. In new load, simply plugged into surplus capacity and only paid a monthly bill, but surplus grid capacity is now gone. New data center load now means new infrastructure, heavy capital, long lead time and rate payers who will not fund or subsidize capacity for a corporate customer.

The faster, cheaper, more predictable path for that customer is islanded on-site power. But that solution demands either a capital budget, most end customers do not have or a financing partner behind the power purchase agreement. And negotiating bespoke terms across financiers, developers, operators, OEMs and end users is complicated and slow. So we kill that friction with strong financial partners who provide our customers' project capital, Brookfield, anchors that financing shelf. We formed the partnership last fall at $5 billion.

9 months later, in June, Brookfield expanded its commitment fivefold to $25 billion, one of the largest and most experienced infrastructure investors in the world evaluated our technology, our delivery record and our pipeline back this with $5 billion, watched us execute and then multiplied that backing by 500%. Capital of that quality and quantity does not follow letters of intent, MOUs or press releases. It follows performance happy customers and firm bankable orders and Brookfield is not alone.

This quarter, industrial development funding, who has previously funded Bloom deployments, partnered with Oaktree, MUFG Bank and Morgan Stanley to fund Bloom deployments cumulatively bringing their total commitment to $2.6 billion and more financing partners are in the wings. Gigawatt demand needs giga dollars of capital, we arranged that in advance. Next friction point, community. Communities have learned they cannot live with combustion. However, we also learned they can live next to Bloom Energy servers, no combustion, negligible air pollution compared to turbines and engines, negligible water use. A power server that is aesthetically appealing and runs quieter than air conditioning equipment.

No construction project is [ Nebius ] proof, but communities welcome Bloom and that is proving to be a real competitive advantage. Our customers can get air permits with our technology faster than they can with a combustion alternator. And every month saved on permits is a month closer to power availability, which brings me to speed because time to power is really time to token revenue. Bloom is increasingly seen as a solution that eliminates the power availability friction point for AI.

Ships without power our inventory, not intelligence. Grid operators got years, time lines on which billion-dollar compute clusters go obsolete in a warehouse. Legacy suppliers celebrate backlog stretching to 2029 and beyond. We think a 4-year backlog is not a trophy, it's a concession of constrained supply. By contrast, Bloom meets the time critical needs of our customers and delivers power in months.

Since the beginning of the year, we have been continuously adding to our American manufacturing capacity in copy exact increments and will continue to do so ahead of committed orders. And our speed is dependable because we have built a resilient supply chain with broadly available materials by design, multiple qualified suppliers across multiple countries for every critical input, inventory ahead of the ramp, relationships forged over 2 decades. No single supplier and no single country determines our destiny. And every part of our supply chain is prepared to scale with our growth.

When the business case is measured in months of AI compute, the fastest dependable path to power wins. We are on that path and nobody else is close. None of this is by accident. We build scale into our model years ago for a simple reason. If we were right about the market need for on-site power, the company had to be able to scale rapidly. Now you're seeing that play out in practice, capital, community permitting, speed, remove all four frictions and the market renders its verdict supplier of choice. I chose those words deliberately because customers are choosing customers who had ordered combustion turbines and reciprocating engines, cancel those orders and choose Bloom.

Nebius did it this quarter. Existing customers are coming back with expansion opportunities and the largest infrastructure investors in the world are underwriting our deployments at scale, trusted partner to customer, to the community and to capital, the market has noticed customers now come to us late in their development process and ask Bloom to step in as the primary on-site power solution. The entry point differs, the outcome does. Not once they see our capability, our execution and the total value we deliver over the life of ownership, the conversation expense from one project to a portfolio.

So let me be very clear. Bloom Energy is not dependent on one customer or one project. It is multiple customers and multiple projects across every stage of development. And because our copy exact lego block servers redeploy from site to site painlessly and like bespoke traditional equipment., every project in the portfolio is fungible. Diversity, fungibility and nimbleness allows us to navigate the fast-changing AI landscape. Consequently, we have visibility and conviction in our growth trajectory for 2026 and beyond.

Let me close with how we run this company. Because I know what's on your minds. Will AI investment keep growing at the breakneck pace? Our engagement with customers suggests that the pace of investment will not just continue, but accelerate. However, I do not know for sure, and I will not insult you by pretending to Henry Ford, could not control that America wanted to dry.

He controls the cost, quality and availability of the Model T. Like Ford, we are focused on managing the controllables. We drive cost down every single year, while others benefiting from these build-outs talk only about raising prices. We innovate continuously to better serve our customers and meet every commitment we make. We are experiencing the tailwinds of rapid TAM growth and increasing market share simultaneously, and we are grateful we can meet the market's demands. At the same time, we are building a durable advantage by earning the trust of both our customers and the communities.

With that, let me hand it to Simon to take you through the numbers. I'll rejoin you for Q&A. Simon?

Thank you, K.R. Good afternoon, everyone. This is my second earnings call as Bloom's CFO and my first with a full quarter behind me. In April, I told you why I joined Bloom, belief in the mission, the architectural shift in power, the quality of the team, the chance to help build a generational company. 3 months in, that conviction has only strengthened. The demand environment is robust, and we are neither gated by capacity nor our supply chain.

The operating discipline is real. The factory improved every week and cost out is a rhythm here, not a program. And in finance, we have a strong foundation, and my focus is on scaling our systems and processes to keep pace with our rapid growth. Before the results, let me come back to the Brookfield announcement K.R. discussed, a fivefold expansion of our strategic partnership as Brookfield increases its framework finance Bloom power projects for AI infrastructure.

A commitment of that size from one of the world's largest infrastructure investors is significant on its own. It is also the best way into our revenue model. Because you cannot fully appreciate what that partnership does without understanding the model behind it. And given our shareholder base has grown considerably this past year, let me take a minute to walk through how a Bloom deal works.

Every Bloom deal starts with a contract with the end customer, the party that will actually use the power. That contract takes one of two basic forms. Either the customer buys the equipment outright, which is a CapEx sale or the contract for power or capacity without owning the equipment. As our 10-K described, the second form comes in a few flavors. A power purchase agreement price per kilowatt hour, a capacity agreement or an equipment lease price on installed capacity.

Economically, they all work the same way. The customer pays over time rather than owning the assets. In any case, every contract carries committed commercial operation date. Since most customers choose to pay over time rather than own, we bring in a financier, an institution like Brookfield that purchases the energy service from Bloom owns them and deliver to the end customer under the contract we originated. So when you see the word customer in our filings, it can mean either party. The financier that buys from us and appears in our revenue and concentration disclosures or the end customer whose demand created the deal.

The IDF partnership that was recently announced is another live example of that model. And it was a meaningful contributor to the quarter. Mechanically, it is our standard structure. Nebius has signed the offtake and IDF, an independent third party is purchasing the energy server on cash terms against identified site and delivery schedules. One feature of this model worth understanding, especially for newer investors, is that large campus deliveries are lumpy.

One or two customers could lead our revenue in any given quarter and the customers rotate a different projects reach their delivery windows. The revenue can look concentrated in any single quarter that concentration reflects delivery timing, not the composition of our backlog. The backlog spans multiple hyperscalers, neoclouds, colocation providers and commercial and industrial operators. And our contracts carry payment security appropriate to the transaction side.

Now to the results. As a reminder, I will focus on non-GAAP adjusted metrics. A full GAAP to non-GAAP reconciliation is in the press release and the supplemental deck on our IR website. Revenue was $1.065 billion, up 166% year-over-year and 42% sequentially. This was another record quarter for Blue and the first time we exceeded $1 billion in quarterly revenue, reflecting accelerating data center deliveries and disciplined execution, converting signed demand into revenue. Product revenue was $935 million, up 215% year-over-year and 43% sequentially and represented nearly 90% of total revenue in the quarter.

Gross margin was 34.3%, up 604 basis points year-over-year. The improvement reflected both favorable mix and margin expansion across both product and services. Product gross margin was 37.2%, up 193 basis points sequentially and up 291 basis points from Q2 of 2025.

A word on price. We deliver value, not a commodity kilowatt. Our customers are paying the time to power and what the solution can do, whether that is following the load profile of an AI campus for being ready to carbon capture and our pricing reflects that. And on cost, we continue to drive product cost down across materials, labor and overhead, and we are doing that while ramping up our capacity and adding new capabilities. Services margin was 22%, up 977 basis points year-over-year and our fifth consecutive quarter of double-digit service margin. Service revenue is recognized ratably, net of guarantees, while the service costs are booked as incurred. So the timing of fleet maintenance, replacements in particular, moves the margin quarter-to-quarter.

Underneath that timing, margins have now reached the 20%-plus level driven by fleet performance, longer stack life and scale, and we believe we will sustain them there over the long term. Blended gross margin will move quarter-to-quarter, driven by the pricing mix of projects delivered in any given period, deliberate trade-offs between cost optimization and expediting where the customers' time to power is worth more than the incremental cost and the service maintenance timing I just described.

Across all of it, we continue to feel good about our full year gross margin outlook of approximately 34%, which we revised up last quarter. Operating income was $240 million, up 737% year-over-year, and operating margin was 22.5%, an expansion of approximately 1,536 basis points. Adjusted EBITDA was $253 million, approximately 24% of revenue. Non-GAAP diluted EPS was $0.78, and GAAP diluted EPS was $0.62. The increase in profitability reflects both higher volume and substantial operating leverage as the business scales. To that point, the operating leverage in these numbers is notable. This is structural, not a 1-quarter effect. Revenue grew 166%, while operating expenses grew just 48% and the mechanics behind that should persist.

The leverage comes from how we have built our cost structure. Our R&D base and our G&A infrastructure are largely fixed against a rapidly growing revenue base. So each incremental gigawatt of deliveries carries little incremental over. We run our support functions the way we run our factories, using automation and analytics across SG&A, service operations and supply chain. So those functions grow with technology rather than people. We will continue to invest across the business, including G&A and R&D, but we expect operating expense growth to remain well below revenue growth which should drive continued operating margin expansion.

Cash flow from operations was $226 million, an increase of $439.5 million from the same period last year driven by profitability and favorable working capital performance. Free cash flow was $175 million, and we ended the quarter with $2.7 billion of cash. Generating significant operating cash flow while growing revenue at this current rate truly reflects the strength of the underlying business and the working capital discipline of the team.

Now turning to guidance. On the strength of our year-to-date performance as well as the visibility and confidence we have for our second half, we are raising our full year revenue outlook to $3.9 billion to $4.2 billion. At the midpoint, that represents 100% growth over 2025 revenue of just over $2 billion. Our outlook is built bottoms up in 2 layers. The foundation is backlog conversion, signed commitment delivering against customer site readiness dates.

The second layer is in-year bookings. We intentionally reserve manufacturing capacity for time to power customers who need power in months and for whom we expect to book and convert into that capacity at rates consistent with recent experience. On gross margin, we are holding our full year margin rate at approximately 34% on a non-GAAP basis. I'll also use this as an opportunity to make 1 point on how we run the business. When we have to choose between protecting a point of margin in a given quarter and expediting to deliver an order quickly to a customer who is going to be with us for years, we are going to prioritize the customer and the long-term strategic value of that relationship. Time to power is what our customers value most right now, and we are going to keep delivering against that. Over a full year, that discipline is fully consistent with the margin rate we are guiding to.

On operating income, we are raising our full year non-GAAP operating income outlook to $800 million to $900 million. And at the updated revenue midpoint, this implies an operating margin of approximately 21% -- that is a material step-up from the $425 million to $450 million operating income guide at the beginning of the year, which represented a 14% margin at the midpoint. This is the operating leverage I described earlier flowing directly through the model. Full year non-GAAP diluted EPS outlook is now expected to be $2.55 to $2.85.

Finally, a word on how to read our guidance. Demand in the AI business does not follow the traditional selling cycle. Today, we see the demand, we book the order and when the customer is ready, we ship the equipment to both the level and shape of our outlook come from the same inputs, which are signed commitment and their schedules, our capacity and our time to power pipeline. What was one seasonality is now simply delivery timing based on customer readiness.

To conclude, this was a milestone quarter. We exceeded $1 billion in quarterly revenue for the first time, delivered record profitability, generated strong cash flows and raised our full year outlook. We are executing with discipline against the demand environment that keeps getting stronger.

With that, operator, we are now ready for questions.

[Operator Instructions] Our first question comes from Mark Strouse, JPMorgan.

K.R., I want to go back to your comment about all of the major U.S. hyperscalers and over a dozen other operators are now validated and approved to use your technology. Are you able to talk about a bit more -- how many of those are actively using your technology today? And how many of those are kind of in your backlog or in your near-term pipeline? And then I've got a quick follow-up.

Mark, as you very well know, we let our customers speak about the deployments and what they do. What I can tell you is the combination of the three things that you spoke about, customers already using it, customers who booked orders and we have shipped units to them for whom the power will be delivered is in construction and customers who have given us definitive agreements. They fall in that category.

We are not going to split that up, but it is all the major, as you pointed out, it's all the major U.S. hyperscalers and over a dozen of the neoclouds and the ecosystem around it, colocation partners. That is all true. We don't break it up. But again, let's just take a moment to think about this.

Nine months ago, we announced our first direct hyperscaler customer and said, we want to enter into this market and do what we did in C&I and become a standard. It took us 10 years to do that previously for commercial and industrial. And I would have -- I can tell you, I never thought within 9 months, we will become the standard. It just speaks to our value proposition in this entire industry, okay? This is not a faster horse. This is a car. And that is why this is happening, and this is not reversible

Okay. Very helpful. If I can sneak one more in. I know going back to the last call, you don't want to get into specific comments on capacity anymore. But just maybe just relative to the last call, I mean just given your comments today several times about things accelerating, is it safe to assume that you were -- maybe your time line or your magnitude of capacity that you're planning is accelerating as well?

Yes. So here's how we do the capacity planning, right? It is very clear to us based on our commercial pipeline and our commercial orders. Then our customers need their products and when they are ready to turn the units on. And as you understand, very clearly, depending on whose reports you read, somewhere between 30 and 40 gigawatts of new AI data center capacity is going to be turned on in 2027, right? And they are in various stages of development. They're all greenfield.

And we have a very sophisticated algorithm that we use to figure out how many of those projects are going to land when. And luckily for us, unlike everybody else, our units are fungible. When they're on a truck, we can redirect them to a different site if we need to. And that's how fungible our systems are. So based on that, we can predict what our capacity needs are going to be such that we never become the bottleneck to the customer. And I can tell you, as we sit here today, we are confident of keeping that promise for everything we have in the order book and everything that we see coming forward. That's all we're going to comment on our capacity. Capacity is not going to be a constraint as we see right now.

Your next question comes from Chris Dendrinos, RBC Capital Markets.

Yes. And coagulations on the strong quarter. I wanted to ask about the supply chain here. And I guess, when you're having conversations with hyperscalers, what are they asking you and what are you telling them or responding to them to the confidence that you won't be the bottleneck in terms of delivering on time?

That's a very good question. Look, you are raising something very important. These are extremely sophisticated consumers and customers. So the amount of diligence that they go through is not just about our product, not just about our performance, not just about our economic value proposition. They want to understand where we are with respect to committed orders, where we are with respect to new orders that can come online and look, most of them, if you talk to them, are not just signing up for a single transaction. They are signing up with us for the future.

They want to be a strategic partner with us for the future. They want to understand as they share confidentially with us what their capacity expansion plans are and ask us if we can meet that. And we have to walk through with them under NDA in great detail and convince them that we will be able to scale. That's when we get validated. So to that previous question of what does all that validation mean? That's the process we go through. It's a pretty rigorous process. It's a pretty rigorous process with each of those customers.

Got it. And then, I guess, maybe fair to say that you go through that same process with IDS and Brookfield. And then maybe just as a follow-up here on the Brookfield topic. You expanded that partnership by $20 billion. How should we think about the timing to execute on that? Is there sort of a window to which you all think about being able to execute on $20 billion of transactions?

Sure. Great 2-part question. You snuck into one question. So -- but I'm happy to answer both of them. They're very relevant, right? The first part is, look, at the end of the day, the financial customers take title of our equipment. When they take title of our equipment, it is not just about can they base it in service. They care about how will it operate through the entire period to be able to deliver.

So they go through additional layers of our ability to perform, our ability to be available, our ability to be available and can we upkeep that equipment for the time that is needed based on the financial model that they came up with, so they can actually get their returns. So it is two additional layers of detail that they would go through in their diligence and they go through the same other process that we discussed. And even in today's world, $20 billion is a lot of money.

So very obviously, they go through this process deeply with us. And again, remember this, they didn't come into it in first step. They put the $5 billion in, they watched how we perform. They watched what we are capable of. They watched the execution. And they spoke to several of our customers, the oldest of them now being more than 15 years our customer to understand how we perform and how satisfied they are. Happy customers is absolutely essential. So it is our strength of all that, they come and invest.

As to the timing, look, think of this as a financial shelf. This shelf is now available. The pace at which it will get used up is going to depend on the uptake. Those funds. I think I can again say it's very similar to us becoming the standard for AI in less than 9 months. I wouldn't have predicted 9 months ago when they invested $5 billion. We'll come back that quickly for the next $20 billion. And this just speaks to the pace of acceleration both in AI and in our business. Thank you.

David Arcaro from Morgan Stanley has the next question.

Appreciate you taking my questions. So there have been some project development challenges that have gotten headlines at a few large projects. I was wondering if you could characterize your financial exposure to project delays, any contractual protections that you typically have in place and alternatives that you may work on with customers.

Yes. Thanks for the question. Look, I think we don't comment specifically on individual projects, as you know. But I take a step back and just think about how we contract, right? Our contracts are structured with master service agreements and there's flexibility for us to deploy this equipment given its copy exact nature to various customer projects.

With that said, we have strong protections with our contract? And then ultimately, our financial years also need those same protections, right? So as you think about the way these contracts flow to the extent that there are any project delays, the end customer is able to redeploy that equipment to other projects. But ultimately, the finance year is on the hook to take delivery of the equipment from Bloom.

And the other thing, extremely important point to talk about in terms of projects because it would be on the minds of people. We can tell you when we gave you this guidance and we upward revised it, right? The 2026 revenue guidance is not dependent on any single project. Again, we have a sophisticated algorithm, and we expect certain projects to push out certain projects to come in, certain projects to just come out of the blue and absorb it in the same year as we described in our script.

So we take all that into account when we give the guidance. So construction projects, as long as construction projects have existed, I'm sure there have been delays. Okay. I'm not a historian, but that's what I would expect. We should bake that in, but that's not going to affect our revenue guidance because we have sophisticated all of them to figure that out for the year. So our 2026 guidance is not going to have any dependence on any single project.

Yes. Understood. That's helpful. And then K.R., I appreciate your additional comments on the supply chain and your confidence there. I was wondering if you could address your access to scandium, which has gotten a lot of attention. Is there any way that you might characterize your use of scandium, how much you see as available supply and stockpile, et cetera?

Look, I think we have put out a detailed blog on this topic and we have filed an 8-K. The three takeaways for you all as investors to understand are the following: there is enough scandium on the planet that can be recovered economically viably to part of the planet. That is what's available on the planet. We have visibility currently based on what we are working for, for 25 gigawatts of like deployments. And so -- and we are not dependent on China. Those are the statements we made. That's all we are going to say about it. Everything else is proprietary to the company.

The next question is from Nick Amicucci, Evercore ISI.

Just curious, so obviously, impressive raise in guidance on a strong 2Q. But the free cash flow guide was pulled, just wanted to get some context around that and just considering that you have $2.7 billion worth of cash on the balance sheet, just thinking through kind of capital allocation.

Yes. Nick, thanks for the question. So just to level set here. The company has historically included a supplemental presentation that has included some metrics that are not formed guidance. And so we've just really aligned now the presentation to where we truly guide. But to come back to your question on cash, which I think is a relevant one, we see significant conversion from operating income down to free cash flow. As you think about where we started the year with an op income guide of the midpoint of $450 million with $200 million of CFOA.

We raised that to $675 million, and now we're looking at $850 million at the midpoint. So there's about $175 million raise versus our prior op income guidance, and we see 100% of that dropping down to CFOA. So you should kind of think about $375 million plus as being our new baseline. But as you know, kind of we don't provide formal guidance as part of our earnings release, but I just want to make sure that you get the comfort that we see strong conversion here.

Got it. Got it. No, that makes perfect sense. And then as we think about just kind of I guess, longer-term AI demand and just kind of why the CapEx from every hyperscaler right now is so high when we think about where the returns are actually going to be generated. Just as we think about that and think about your backlog and kind of the conversations you guys are having K.R, are you starting to have any conversation just along the lines of inference reasoning? Or is it still really just kind of speed to power the -- on AI training.

Yes. So it's both. It's absolutely both. Look, time to power is super important, number one, okay? And I think -- let me try to explain time to power in a slightly different way because for many of the analysts who look at utilities and power as well as some of our investors, it's important for you to understand a full stack AI provider that is responsible for everything in the data center financially, a 1 gigawatt data center in one single year, depending on the nature of the AI customer will deliver between $12 billion and $24 billion in revenue per year.

So you pull in power for them within a month, which is the tall pole that is $1 billion of revenue that they would not have had on a 40% to 50% gross margin and a 20% to 25% net margin. So among the 35 to 40 gigawatts that needs to get deployed next year, take a guess on how much of that those projects will be delayed because the power provider is not able to provide power on time, okay? We are the place to go. for time to power. And the math is you don't even have to do the math if we can provide that. This is why time to power is important for the large data centers.

Now as influence comes along, if the transmit and distribution infrastructure in the country is having difficulty doing transmission, building highways, imagine how difficult it will be for them to upgrade distribution, which is surface streets. That's where inference power is going to be needed. Bloom is ideally suited for that. You cannot put a gas turbine in the middle of Manhattan. So we see both opportunities as extremely robust for us, not this quarter, not next quarter for years to come.

Your next question is from Ben Kallo from Baird.

I have two. I don't know if you guys think about competition and supply demand curves out there. But maybe if you do, could you give us your thoughts about where we sit just in overall new capacity coming online, whether it's reciprocating engines or it's combined cycle turbines or other versus your decision process? And then I have a follow-up, which is kind of at a higher level.

I think given the huge supply demand gap right now, I think every single technology that can provide power quickly for the next few years is going to help play. Let's just start with that. So if engine makers, turbine makers increase their capacity, there's going to be a need for it. If Bloom increase its capacity, there's going to be a need for it. But let's fast forward and think through competitively at a point, if a single customer has to choose between a turbine engine and a fuel cell, okay?

The first thing that matters is not LCOE, which is just an absurd construct for on-site power. It is the total cost to tokens, the total cost of power to token revenue. Bloom's ability to provide 800-volt DC power, Bloom's ability to provide reliability without overbuild. Bloom's ability to be able to locate inside a city outside of city because we don't pollute the air. Bloom's ability to get a permit, okay?

None of the other competition, there is not a single commercial vendor today who can provide that total value proposition other than Bloom. So what we do then is we don't obsess on the competition. we obsess on the customer, okay?

Just maybe from a commodity model or Chinese open source models that have I think people worried or uncertain. Can you just talk to us about if you view that as an opportunity, either of them or a threat going forward?

Look, whether it is the Chinese, whether it's the American labs, it just doesn't matter. It is a -- I would say, as a technology optimist, to me, it's a given that token cost and token efficiency and tokens ability to do things are all going to improve. In that cost will get better, cheaper, efficiency of the token will get much better what a token is able to do with respect to productivity will keep increasing. All of them will happen.

That means the token price point will come down, but the total token usage will go up like crazy because at G1's paradox. And when that happens, you need more power, not less power. So if anything, this is going to accelerate, if anything, whatever we are predicting on AI is an underestimate, not an overestimate. Thank you.

And everyone at this time in the interest of time, we will switch to just taking one question. Our next question is from Manav Gupta, UBS.

You built this company -- started building this company 21 years ago, you had a vision of the cell. And obviously, you've come a long way. I'm trying to understand what's the vision of the product for the next 3 to 4 years, coming back to your preliminary comments of what you can control in context of Henry Ford. So where do you see this product moving in the next 3 to 4 years?

Look, I think very clearly, imagine our on-site power, DC is going to be the primary source, whether it's data centers, whether it's anywhere else, whether it's fleet charging of electric vehicles, whether it is a large apartment complexes and microgrids being built for residential complex as DC is where the world is going to go, right, predominantly.

So DC power being generated on site, being able to use the heat to do both the heating and the cooling and then on top of that, on like top of that, decarbonization in my view, is going to become super important, and Bloom is able to do carbon capture better than anybody else, okay? So we're going to be focused on how do we give a comprehensive solution where the fuel is getting utilized at 90-plus percent efficiency, combined efficiency. It is not polluting the air. It's not using water. The same format that a customer gets used to, to power large data centers, is the same format, the same technology that powers your neighborhood store, that powers your neighborhood inference data center because that is the vision, and we want it to be like an appliance that you can plug in and get power. This is where we're going to go.

Next up, Maheep Mandloi from Mizuho Securities.

One question on capacity expansion for the next few years. Like one theme we're hearing from other manufacturers across other industries is just inflation on the CapEx estimates. Just curious on your thoughts on these. I know your manufacturing equipment is different. So how should we think about that now as you expand from 2 gigawatts to beyond that?

Thank you so much for asking that question because that's a significant distinguisher for us, right? Our factories, the return on investment is a few months, okay? We don't come from the whole world. Okay? This is not your industrial age power company. We are relying on the technologies that made consumer electronics and semiconductor devices become better available in larger quantities to everybody on the planet and get lower in cost and provide greater value. That's the model we are adopting. That's the model we're going to follow.

So let them deal with whatever they're dealing with from our perspective. For us to expand capacity, the return on investment is going to be a few months, and we keep adding capacity as long as the demand is there.

The next question will come from Sunaina Ocalan from Bernstein.

Maybe if I can ask about the competitive landscape and just a follow-up on the comments that were made on the call as well as somebody's question ahead of mind. It makes sense. I think what you're saying makes sense, which is on-site power Bloom solution, not having NOx and SOx, not using water is a superior solution. That makes sense and it's clear. How are you thinking about the market share over the next maybe 12 to 24 to 36 months on some of the other fuel cell models that are are targeting essentially the same data center space. So I'm getting asked about molten carbonate, for instance. If you can just provide some color on the fuel cell market share, that would be great.

Look, I think it's for them to tell you how many megawatts and how many gigawatts they can install, okay? It's not for us to comment to you today within the data center space, I would say we would be in the very high 90s in terms of our market share, okay? If like somebody is going to get in and think that they're going to compete with us, competition is a very good thing, okay? Competition makes us hungrier. Competition makes us run faster. Competition makes with paranoid, and we will thrive in competition. So I welcome competition from anybody and everybody.

The next question will come from Colin Rusch, Oppenheimer.

Guess if you get into this a little bit deeper with the data centers and have those time to power advantages. Can you talk a little bit about evolution of your thinking on pricing and target margins for the platform as well as just giving us a sense of how many projects you're selling into where you're displacing a different technology that was originally planned for those sites?

Look, again, we don't think about LCOE price of electricity because we're not a utility, okay? We are a strategic partner to our customers, and we deliver value to them. And based on that value, they should be happy buying from us, and they should be happy allowing us to capture value. So it is not just a sense per kilowatt hour story for us. It's about the added benefits we bring to them. It's about an entire solution that we look at.

And for us, being able to increase these margins are by capturing value is extremely important. And look, when you think about this, right, we are so excited about talking about growth. We forget to highlight something extremely important that's happening in the company, okay? I remember 7 years ago, most of you, the same folks analysts, the only thing you're worried about is our service losses. Okay? That's the only thing you're worried about.

And we kept telling you the technology product is going to get robust and you will see us get to the 20% gross margin that we talked to you about. So we just reported this quarter a 22% gross margin. I want to take this opportunity to give a huge shout out to the team out here that's worked tirelessly to go make that number happen. Think about it. When we went public 8 years ago, minus 21% gross margin, negative 21% this quarter, plus 22%, 43% swing in service margin.

Now, let me focus on one more thing. It's service margin is a financial metric. But the first word in that is service. Who are we serving? We're serving our customers. At the end of the day, it's happy customers that matter. So we are not just achieving that financial metric. If you look at 2025, 80% of of the orders that we booked were repeat orders from customers have given us multiple repeat orders. That speaks more loudly than anything else about how happy our customers are. So very, very proud of that accomplishment.

I truly believe, and I'm going to close with this. I truly believe that, that service part of the business and the service revenue and the margins is a big driver to our enterprise value, and more importantly, how we serve our customers in a proper way. Very proud of that team. I want to give a huge shout out. And if you combine what I just told you, by looking at our backlog, the demand out there, how we fit perfectly to the needs of a future digitized world unlike previous technologies. Look at how we are executing as a team.

And then combine that all together to see the trust that we are building within the communities we operate and the companies we serve. I have a lot to be grateful for and a lot to be appreciative of a great Bloom team that has done a phenomenal job.

Ladies and gentlemen, this does conclude today's conference. We would like to thank you all for your participation. You may now disconnect.

Ladies and gentlemen, thank you for standing by. My name is Abby, and I will be your conference operator today. At this time, I would like to welcome everyone to the Bloom Energy First Quarter 2026 Conference Call. [Operator Instructions] And I would now like to turn the conference over to Michael Tierney, Vice President, Investor Relations. You may begin.

Thank you, and good afternoon, everybody. Thank you for joining us for Bloom Energy's First Quarter 2026 Earnings Call. To supplement this conference call, we furnished our first quarter 2026 earnings press release and supplemental financial information with the SEC on Form 8-K and have posted these materials, which we will reference throughout this call to our Investor Relations website.

During this conference call, both in our prepared remarks and in answers to your questions, we may make forward-looking statements that represent our expectations regarding future events and our future financial performance. These include statements about the company's business results, products, technology, customers, new markets, strategy, financial and competitive position, investments, liquidity and full year outlook for 2026.

These statements, which relate to matters including time to both power and market with standard for on-site power cost efficiency, capacity expansion, innovation, affordability and community acceptance as we look to keep pace with the rapid evolution of our markets are predictions based upon our expectations, estimates and assumptions.

However, as these statements deal with future events, they are subject to numerous known and unknown risks and uncertainties, as discussed in detail in our documents filed with the SEC, including our most recently filed Forms 10-K and 10-Q. We assume no obligation to revise any forward-looking statements made on today's call.

During this call and in our first quarter 2026 earnings press release and supplemental financial information, we refer to GAAP and non-GAAP financial measures. The non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP.

A reconciliation between the GAAP and non-GAAP financial measures is included in these materials, which are available on our Investor Relations website. Joining me today are K.R. Sridhar, Founder, Chairman and Chief Executive Officer; and Simon Edwards, our Chief Financial Officer. K.R. will begin with an overview of our progress, and then Simon will review financial highlights for the quarter.

After our prepared remarks, we will have time to take your questions. I now turn the call over to K.R.

Good afternoon, and thank you for joining us today. Bloom delivered a record first quarter. Revenue, gross margin and operating income all came in materially above our prior outlook based on what we are seeing across the business, we are also raising our full year guidance and raising it materially.

We will walk through the numbers in a few minutes but first, I want to talk about what is happening in our market because the headline numbers as strong as they are, are a lagging indicator and don't convey the whole story. Fee at Bloom are ushering in the era of digital power for the digital age. Now the marketplace is recognizing and embracing our proposition of clean, reliable on-site power that is community-friendly and deployed at the speed of AI.

Bloom is rapidly becoming the standard and go-to choice for on-site power. Last night, Oracle announced a new power paradigm for Project Jupiter, a multi-gigawatt AI factory to be built in New Mexico. We are thrilled to partner with Oracle and applaud them for their visionary leadership. This up to 2.45 gigawatt power block will replace Project Jupiter's previously planned gas turbines and backup diesel generators with Bloom Energy Servers.

It will be 100% bloom. When completed, it will be one of the largest islanded microgrid power facilities in the world. Oracle pivoted to Bloom only solution for 2 main reasons: first, be a responsible corporate citizen and partner by being responsive to resident concerns about air quality, water use, noise and increasing electricity rates.

Second, to stand up their grid independent and clean AI factory with even greater reliability and speed. Bloom is the cleanest commercially available on-site power generation option for such data centers and the most water efficient. Even Blooms community-friendly attributes, Oracle should be able to energize the campus materially faster than any other available alternative solution in the market.

At a time where every quarter of delay translates into hundreds of millions in foregone AI revenue and loss of competitive advantage. Speed of powered infrastructure development is the difference between leading and following. Becoming the sole power provider for Project Jupiter is a milestone for Bloom, but it's not going to be a one-off project, where Oracle is going is where the broader market is headed.

On our Q3 call, I described our playbook for becoming the standard. In each vertical, we established credibility with the lighthouse customer then build on that success with other Tier 1 customers. 2 quarters later, that's exactly what's happening across the AI ecosystem. Oracle is rightfully getting headline attention today.

But well more than half of our current data center backlog comes from other hyperscalers, neo clouds and colocation providers. just like the Oracle Jupiter project, these microgrid installations will use no grid, no dirty diesel generators for backup. No battery banks for load following. No engines, no turbines, just bloom and Bloom alone.

We are continuing to engage with more hyperscalers and new clouds by signing new contracts and slot reservations and working with them to evaluate many new opportunities. Our pipeline today is diverse and robust in the AI segment. Parenthetically, let me also remind you that this is a recent repeat of our C&I business playbook.

That segment is also experiencing strong demand is diverse and continuing to grow. I want to give you a perspective on why we are experiencing the hyper growth because it will shape how you think of Bloom going forward? For over 25 years, we built this company around the conviction that clean, reliable, affordable on-site power would become essential to a digital world.

The market is now validating that vision at scale and AI power demand is simply accelerating it. Time to power has gone from a procurement consideration to an existential necessity. The company is driving the AI transformation are raising against each other on the one hand. And on the other, bumping into the bottlenecks comment to building conventional infrastructure such as permissions, permits and community acceptance.

The winner will be the one who can grow and deploy faster and on the schedule, the market demands. You see that's a different game than the one the legacy power industry is set up to play. Their model is industrial. Long cycle times, capital heavy capacity additions, product improvement measured over decades rather than quarters, our model is different at every layer we innovate and improve continuously, be it in our technology, in our product in how we manufacture in our capital intensity in how we deploy in how we operate and service our systems and in time to market.

That is what allows us to deliver double-digit cost reductions year after year expand capacity with materially less capital than industrial era players and meet our customers' schedule needs. Our differentiated and unique operating rhythm and mindset will be obvious to you if you visited our factory floor.

It's a state-of-the-art production facility a busy construction zone and a buzzing innovation hub. We are manufacturing product on schedule to meet customer needs, adding lines and expanding capacity to meet growing demand and innovating to reduce cycle times, space needs and costs. Product manufacturing, capacity expansions and innovation, all occurring concurrently all the time, all under the same roof and all with factory floor team members and engineers working as 1 team for 1 common purpose.

To be better tomorrow than we are today and keep marching towards the north star of maximum entitlement. This is an example of our operating model. We call it the Bloom way. As a result of this approach, the contrast and outcomes is simple. Their supply to current orders arrive only in 2029 or later, irrespective of the customers' needs.

Hours arrived this year or the next or whenever the customer is ready. Based on demand profile, we have now shifted to adding capacity continuously. Hundreds of megawatts a quarter as opposed to lumpy one-off additions to be completed in a year's time. How we think about and execute on capacity addition is one of the clearest ways to see what makes Bloom different.

The traditional power industry has been the past 2 years, celebrating its backlog that is 4 and 5 years out. Backlog at that scale and time frame in the age of AI is a result of their constrained supply.

At Bloom, we see it differently. Our ability to expand capacity is our competitive advantage. We want to rapidly build capacity, build product help build productive AI factories to help build commercial and industrial facilities and help build our economy, not just be satisfied with simply building backlog.

Our current manufacturing footprint will allow us to deliver 5 gigawatts of product annually. We will expand to that capacity and meet the delivery dates needed by our customers. In other words, today, we are not order constrained and not capacity constrained. The pace of our revenue growth is decided by how fast our customers can build their greenfield sites, not how fast we can power them.

We will never be a bottleneck to our customers. We built our business around that promise. Going beyond the 5 gigawatt capacity, our supply chain and manufacturing strategy and planning allows us to build that capacity significantly faster than any other option in the market using our copy exact model. We will strive to bring power to our customers faster than they can stand up their greenfield facilities. We were able to make that promise because we invested deliberately ahead of demand.

We expanded manufacturing capacity, built inventory, diversified our supply chain, strengthened our balance sheet and assembled an ecosystem of long-term supply partners that scales with us. Given our low capital intensity, those investments carried materially lower risk for shareholders than they would have for an industrial or a supplier.

They were disciplined decisions made with conviction that this market shift was coming. While our new orders that we are telling you today are news to you, we have advanced visibility and anticipated such wins for months. So we planned out our capacity expansions accordingly.

Our strategy and judicious investments have positioned us to become the standard for both on-site power and time to power. Beyond speed, our architecture creates real flexibility for our customers. Our modular copy exact systems are portable and fungible and meet air quality requirements in virtually all jurisdictions.

If a customer needs to shift deployment from 1 site to another, our master services agreement is structured to enable that. With the master service agreement, our hyperscale customers have the geographic flexibility to move a bloom deployment from 1 site to another based on a speed up at 1 site or a delay in another.

Bloom moves with the customer to the location where the GPUs are ready to convert the power to tokens of intelligence and revenue dollars. Unlike a traditional power plant, our platform is also a different kind of neighbor in a community. We are community friendly. As more on-site generation gets deployed to support AI and industrial growth, communities care deeply about what kind of infrastructure shows up next door.

Bloom preserves local air quality, we do not combust and pollute the air like conventional technologies. We use minimal water edge startup and none during normal operations. we acquired compact and efficient with land use. We integrate well with environments rather than disrupt them and become an ISR. As permits and permissions become the gating factor for AI infrastructure, community acceptance matters increasingly.

Our fully landed grid-independent one-stop full stack power solution does not raise the monthly electricity bill for community residents and brings them economic development without compromises. The cost equation has also shifted in our favor. We have spent years driving down product cost while improving performance. That work is meeting the market at exactly the right time. our energy servers are now cost competitive with grid power in most U.S. markets and with off-grid alternatives in nearly all markets.

With over a decade of double-digit cost reductions, we remain the only on-site generation solution with a sustained downward sloping cost curve. As affordability of power becomes a national issue, we expect to become the solution of choice from that perspective also. Bloom delivers a value proposition built on the principle of and not all, customers can have the power that is clean and reliable and fast and affordable.

Now to our outlook for the year. To say that the commercial landscape is fluid and dynamic would be a massive understatement. The strength of the quarter and the commercial momentum we see across the board gives us conviction and confidence to raise guidance materially. We are raising 2026 revenue guidance of $3.1 billion to $3.3 billion to $3.4 billion to $3.8 billion.

At the midpoint, that takes growth from 60% year-over-year to 80%. We are also raising our gross margin outlook from 32% to 34% barring any global shock or exogenous factors. You can see, we are prioritizing growth and profitability in equal measure. Now I want to introduce Simon Edwards, who recently joined Bloom as our Chief Financial Officer.

Over the past year, we have been deliberate in our search. It was important to us that we not only find the right CFO for Bloom today, but the right leader and business partner to help bloom scale for the future. Simon brings a rare combination of capabilities. With the systems engineering background, he has built disciplined operating models and scaled manufacturing operations for complex systems as CFO of leading software franchises, he has applied a digitally native approach to building businesses, leveraging data and analytics as competitive advantage and employed automation for speed and efficiency.

His time at Grok has given him a front-row seat to the explosive growth occurring across AI. All of that translates directly to where Bloom is headed. I also want to thank [indiscernible] and the finance team for their outstanding work in supporting the business without missing a beat during last year. Their performance speaks to the depth of the bloom talent at all levels. I'm proud.

Finally, to the Bloom team, thank you. What you've built over more than 2 decades is meeting the market at exactly the right moment. You believed and always knew that an inflection point would come. None of what we see today would be possible, but for your faith, dedication, diligence and discipline, much gratitude.

With that, Simon, a very warm welcome, and the mic is yours.

Thank you, K.R. I appreciate the kind words today and the warm welcome that I've received here at Bloom over the past couple of weeks. I'm excited to be part of the Bloom team and to be speaking for the first time on a Bloom earnings call. I chose to join Bloom for a few reasons. First, K.R. talked about the architectural shift driving a large TAM with increasing momentum. Having seen the powerful tailwinds around AI infrastructure and electrification, I recognize very real bottlenecks in power availability.

Bloom is uniquely positioned to address that challenge with a long-term opportunity that extends well beyond AI. Second, Bloom is a Silicon Valley innovator, solving an industrial problem. I was drawn to balloons visionary leadership and the depth and quality of the leadership team. There is a clear strategy, strong alignment and a mindset focused on building something enduring that starts with K.R. and permeate through the entire organization.

And third, this is a chance to help build a truly generational company, one that can capitalize on long runway for growth and create long-term value for customers and shareholders. Since joining 2 weeks ago, I have already been impressed by what I have seen. The team is highly engaged and motivated. The demand environment and pipeline are exceptionally strong, and there is a clear bias towards the results. turning that demand into delivered systems, cash flow and sustainable performance.

In addition, the sense of mission is clearly apparent among Bloom's employees. Many of our employees have been here for 10 to 15 years, long before AI was a common phrase. These employees stayed here because they believe in the Bloom mission. To make clean, reliable energy affordable for everyone in the world. This is a driving force behind everything we do here at Bloom and the mission I'm excited to be part of.

Moving to our numbers. I will discuss our Q1 financial performance and make a few comments about what we expect in 2026. Highlights include record Q1 revenue with year-over-year growth of more than 100%, continued year-over-year gross margin expansion and record Q1 cash flow. As a reminder, I will focus my discussion on non-GAAP adjusted financial metrics. For a reconciliation of GAAP to non-GAAP, please see our press release and the supplemental deck on our website.

Revenue for the quarter was $751.1 million up 13.4% year-over-year. This is the first quarter of greater than 100% year-over-year growth in Bloom's history as a public company. Product revenue was up both year-over-year and sequentially. The reaching an all-time high of $653.3 million for the quarter. Service revenue for the quarter was $61.9 million, up 15.6% year-over-year.

Gross margin for the quarter was 31.5%, up approximately 280 basis points versus last year. Product margins were 35.3%, up 22 basis points from Q1 last year. As we grow, we should see incremental progress on product margins through scale, better absorption of manufacturing overhead and from the continued cost-out efforts across engineering and supply chain. Services margins were 18%, up 13 points from Q1 last year, achieving a double-digit gross margin for the fourth consecutive quarter and profitability for the ninth consecutive quarter. We expect margins for the services business to continue to benefit from both growth and scale and field performance improvements.

Operating income for the quarter was $129.7 million, compared to $13.2 million last year, an increase of $116.5 million with operating margins reaching 17.3%, up more than 1,300 basis points year-over-year. Adjusted EBITDA for the quarter was $143 million compared to $25.2 million last year, an increase of $117.8 million with EBITDA margin expanding by more than 1,100 basis points to approximately 19%.

This margin expansion highlights the significant operating leverage in the model as revenue growth continues to outpace cost growth. Non-GAAP fully diluted EPS for the quarter was $0.44 versus $0.03 a year ago. While we will continue to invest to support the growth ahead of us, I'm impressed so far with Bloom's ability to deliver at an increasing scale while managing costs through both operational efficiency and gaining leverage through technology adoption.

As K.R. mentioned earlier, we are rapidly expanding capacity through our innovative manufacturing model, which allows us to scale in months, not years. That growth requires upfront working capital to support higher production and deliveries. Even with those investments, cash flow from operating activities was an inflow of $73.6 million, positive for the first time in the first quarter of the year, which is typically a seasonally weaker period. This was driven by a step change in profitability, strong collections and customer prepayments to reserve capacity.

We ended Q1 with $2.52 billion in total cash on the balance sheet. Turning to guidance. After a strong start to the year, and anticipating that Q2 revenue should be at least as good as Q1, we are raising our fiscal 2026 guidance to new levels.

We are increasing our revenue projections from the previous range of $3.1 billion to $3.3 billion up to a range of $3.4 billion to $3.8 billion, with the lower end of the updated range sitting above the upper bound of the prior range. This updated guidance represents 80% year-over-year growth at the midpoint and reflects the progress we have made in adding manufacturing capacity, the strength and velocity of our pipeline and the opportunity to continue to prosecute a healthy backlog.

We now expect our non-GAAP gross margin to increase from 30% in 2025 to approximately 34% in 2026. We representing about a 4-point improvement year-over-year to 2 points above our original guidance as we realized the impact of ongoing cost optimization and productivity initiatives. Our non-GAAP operating income expectation is now $600 million to $750 million, acknowledging the higher revenue and margin flow through, but also recognizing that we plan to invest to support the growth for this year and the future.

Our non-GAAP fully diluted EPS expectation is now $1.85 to $2.25. To conclude, we delivered record Q1 financial results, and we are optimistic in our full year 2026 financials being the best in Bloom history. I'm looking forward to working with KR and the entire Bloom team and spending time with our analysts and shareholders. Operator, we are now happy to take questions

[Operator Instructions] And our first question comes from the line of Mark Strouse with JPMorgan.

Maybe starting with Simon. So first of all, congrats on the new role, and welcome to the fray here. You mentioned kind of how you're impressed with the operating leverage in the business. I'm just curious, I fully appreciate you haven't been there very long, but kind of your initial take...

Go-to-market. There's a number of growth factors that we're exploring. Obviously, from a technology standpoint, growth is highly innovative, and we're investing in innovative areas. And then on top of all of that, I think KR has mentioned in the past, cost reduction is in the DNA of Bloom. And so I think really what we're focused on is, a, how do we execute on the projects in the plan right now that deliver on the gross margin expansion that we've highlighted.

Second is, as you look at our updated guide, you'll see there's incremental operating margin expansion baked into the revised guidance. And then finally, as it relates to longer-term guidance here, I don't think that's something we'll provide right now, but continuing to execute on these vectors is something that I know everyone here is very focused on.

Got it. If I could ask one more follow-up to K.R. Clearly, your orders are accelerating here. I'm curious if you can comment on what you're seeing with your service contracts, particularly the duration of those contracts is I think in the past, you've said some of these data center contracts have been somewhere around 6 or 7 years in duration.

I'm curious if you're seeing any change there potentially longer than 10 years or so somewhat similar to your C&I business.

Mark, thank you. And look, I think it's important because some people may be coming in new into the story, we have a 100% attach rate between our product sales and our service. That's the first place to start. There is not a single deal that we do without an attach rate to our service. Even with the data center opportunities, on average, it's 10 to 15 years, somewhere in that range.

And so it's a tremendous source of annuity revenue that we see. And you can see us executing on the margin targets that we have provided. So it's going to be a phenomenally great business for us going forward, along with our product business.

And our next question comes from the line of David Arcaro with Morgan Stanley.

Congratulations on the results here, and a warm welcome to Simon as well from me. But maybe I was wondering if you could touch on the pricing backdrop that you're seeing. I'm wondering if you're seeing opportunities to hold pricing or potentially seeing projects with increased price opportunities in the current environment just where we're seeing -- it seems like all other alternatives are increasing cost to the customer.

So we completely distinguish and think differently about this. at the end of the day, we don't compare our pricing with engines and turbines. It's apples and oranges. We are creating a completely different value for our customer. be it 800 old DC being eliminating all the paraphernalia, the Band-Aids as I've called them to a mechanical solution going to a digital age, be it the amount of overbuild that you need to have when it comes to getting the reliability that you need because it's very obvious, these big projects are not going to have grid backing it up. .

The local rate payer is not going to be providing that reliability for free. And so you bake all that in, we just always focus not on cost, not on price. Obviously, we are going to create margin for the business. And as the first question was asked to Simon, we will focus on that. But at the end of the day, we're going to build our business with our partners. By creating value for them and creating value for us. So we don't look at anybody else's pricing and what they do. Thank you.

Understood. Yes, that's helpful. I appreciate you characterizing that. And I was wondering as you look to ramp up your scale significantly here, could you also speak to how you're seeing the supply chain and its ability to ramp with you? We've seen labor as an example, become a constraint elsewhere. I'm wondering pressure there or in upstream materials?

That's a great question, David. Thanks for asking. And because that's a significant distinguisher between what we do and what other people do. So if you had come to our factory and seen the few hundred people that we have manufacturing our stacks than we were doing 200 megawatts a year.

And if you came at the end of this year when we will be doing almost 10x that amount, the number of employees on the shop store will be the same not almost equal, will be the same. And that is the innovation we bring into the field, knowing that for us, automation and figuring out how to train our existing employees, upskill them as they grow.

And by the way, most of them happen to be the same employees, too. They are upskilled from doing that manual labor to automation. That's why with their hub, as you heard in my script, in the shop floor, we don't talk in harsh stones about bringing automation to remove a particular manual process out because our team members are actively involved in it. And this is the same philosophy with which we are approaching our supply, we have approached our supply chain and are approaching our supply chain because these were custom suppliers built for us in whom we expected that same Bloom way mentality, and we're enforcing it. So the ramps you're talking about are [ pre-seed ] brands. Can you hit some speed bumps along the way, maybe, but are we worried about it or lose sleep or think that we cannot get over those bumps? Absolutely not. We are confident in being able to deliver the promises we make to our customers, not just because we have a very good manufacturing shop. For us, that manufacturing extends to our supply chain partners, and they adopt the same philosophy. Thank you.

Our next question comes from the line of Chris Dendrinos with RBC Capital Markets.

Just echoing the congratulations on the strong quarter and welcome to Simon. I guess my question here is, if I go back last quarter, you had talked about scaling capacity as your customers call in to order that. Now you're talking about continuous capacity increases.

So I'm just wondering if you could provide a bit more color sort of on what's changed here in the past months that changes that approach? And what are you seeing from your conversations with customers to give you confidence to continue to expand here?

Chris, thank you for that question. Look, to say that business is accelerating as an understatement. Okay? We are very, very clearly seeing that demand. And we just don't look at the demand is coming to us at any point in time in isolation. VRA power company embedded in Silicon Valley, and we understand the end-user technology extremely well.

We can get into the basics of what is happening in the field of AI and understand why that demand is going to be there and I can tell you, this is a secular demand that's going to last for many, many years to come. It is with that conviction when we draw to that conviction and we understand.

We talked -- if you remember a couple of calls ago, about people just grasping on the crumbs of utility capacity being available. Those comps have been eaten up. So we clearly see where this is going to go. And we see what we fundamentally see is the following: the amount of demand that is being generated and the rate at which that's growing is significantly faster than what alternative providers of power can create.

That creates a beautiful opportunity for us that we see over many, many years, and it gives us the confidence to be able to say we are now going to continuously grow. So think of Bloom's capacity increase as an analog dial that constantly keep increasing as opposed to some digital step function that happens once in a while.

Got it. And I guess maybe just as a follow-up to that, and that step function comment. I mean is there a step function between going to 5 gigawatts and then maybe going beyond 5 gigawatts. Do you need to see something different from like a customer commitment schedule to add physical footprint? Or do you think about absolutely adding an extra facility the same way.

Yes. That's a valid question. Absolutely. So the answer would be the following, right? As we said there, whenever we made that statement to you, their existing facility was 5 gigawatts. In my script, I talked about we are constantly innovating. I don't know how much more we can milk out of it.

But no matter what we do, we are going to need new factories as we go forward. Bloom was built on the vision of lighting up the planet. Okay? 5 gigawatts a year or 6 gigawatts a year is not going to light up the planet. So we are going to build factories as needed. And that's just going to be a normal course of operation for us and the step functions at which we grow will purely depend on where the market is and where the market needs us.

And our next question comes from the line of Nick Amicucci with Evercore ISI.

Great. and welcome, Simon, look forward to working with you in the future. Quick question for you, K.R. Just kind of piggybacking on Chris' question. So when we're kind of seeing that -- seeing demand and it's not coming through an isolation. Is it fair to say, too, that the vast majority, if not all, of the backlog currently is probably tethered towards your towards like AI training. And then there's conceivably an incremental leg of growth when we kind of think about inference and just the lack of need for air permits and kind of the ease of siting and permitting and so on?

You're absolutely right. Let me tweak your statement in the following way. influence is going to be much bigger than training in terms of total gigawatt need. But it is going to be not concentrated in the multi-gigawatt data centers that you're looking at. And think about this influence by definition, is at the edge, a lot closer to highly dense populations of people and processes. If you're seeing the resistance you're seeing today to a conventional power plant being built in the backyard of a large training data center that happens to be in a small remote town.

Just think about what that resistance would be in a city if you don't have clean solutions. Let me put this in perspective for you, okay? You just heard about the Oracle announcement of up to 2.45 gigawatts. I'm going to use that as an example, not that particular site, take that number. think about a 2.5 gigawatt power block that needs to power a large training data center somewhere.

The obvious example that you would go to would be a large CCGT, a bunch of large CCGT with gas to be able to provide the power. To put it in perspective for the people listening on this call, that is the capacity of the state of Rhode Island in one single data center. And that happens to be, if you use CCGT you will use all the water that all residents used to shower a day in Rhode Island just to power that power plant close to 1 million showers a day. And it will create not from it, air pollution that is the equivalent of all the cars in Rhode Island almost in that one location.

So even in a remote town, you can understand why there's a pushback and why clean is going to be important. If that's how important it is for a large data center, Imagine now for influence where it's going to go. So we see that as a huge opportunity coming our way as we go forward.

Great. No, that's definitely helpful. And then , as we think about -- obviously, there are certain kind of other, I guess, product on competitors kind of coming out with kind of solutions that are more of a bridge power type of solution.

Are there any conversations that you guys are having with kind of your hyperscaler customers or the neo class where it's kind of -- we want to leverage the fuel cell to get up and running speed to power is paramount, but ultimately still feel the need to be grid tied or just given the reliability attributes of your fuel cell offering, is that kind of a moot point?

Earlier in this conversation, they used to bring up the concept of Bridge power with us. And I would smile and always say we're happy to sell you a bridge to a bridge because Superman coming, okay? So today, that conversation is nonexisting.

And our next question comes from the line of Manav Gupta with UBS.

Had somewhat of 2 technical questions and ask them together. While there are other solutions, but they do depend very heavily on battery packs, for load balancing and backup, batteries are expensive, they decay they take place and they generate heat. Your solution with ultracapacitor and high reliability needs minimal battery backup. In some cases, no battery backup. So can you talk about that?

And the second question is, as it is getting clear that cyber and Rubin are the future those building those hyperscalers are looking for conversion parts that can help them go from you have [ 415-volt DC ] to 800-volt DC. Now based on the channel checks we have done, large power transformers, medium board switch gears, centralized rectifiers are all seeing long queues and delays in shipment again, your solution avoids those costs and those delays. So can you talk a little about these 2 factors?

Manav, thank you. Do you want to come work for us? You're making a very good sales pitch here for Bloom. Obviously. Look, this is what you're saying is very true. Here are the 2 things. Number 1 we are purpose-built and purpose designed to provide digital part or digital age. .

Now it is fully understandable as I see it for large data centers to be extremely cautious about introducing any new technology, until it's proven out because the stakes are very high for them. So we had to pay our deals and slowly get in and become a pull the solution out, using AC, using all their backup generation, everything.

But today and the most important point I want to highlight to you from like today's script that you saw from me. It's not just the deal we did with Orca, but we talked about several other projects we're working on, where there is no grid connectivity. There is no diesel backup generators.

There are no turbines and there are no occasions. And like you correctly pointed out, there are no batteries because 100% bloom one-stop solution can solve that for them in our combined solution between our fuel cells and our ultra cats, okay? So that has to start resonating and it started resonating.

Now the next step is for them to go straight to that 800-volt DC.

It is -- as I see it, it's self-evident to me that, that is going to come. It's inevitable that they're going to switch to that. because the world does not have enough copper like you pointed out, the world does not have enough transformers. So necessity is going to force them there. And once they try it, they will not go back on it.

And our next question comes from the line of Ben Kallo with Baird.

Congrats and welcome, Simon. K.R., I wanted to talk just maybe on the demand front and just the different channels, we saw your largest utility deal, and you've had utility deals before. .

Now, obviously, with Oracle and hyperscalers you have repeat customers there. How do you think it evolves where the growth comes from additional hyperscalers? Is there a new channel like midstream gas companies, something like that? And then how do we think about the international side of the business? Like is that kind of a delayed growth area, just kind of lagging what the U.S. is doing here? And then I have a follow-up.

Great questions. So let me be quick and answer those questions in the following way. You're absolutely right. What we are doing in AI right now is truly a rinse and repeat of what we have done in the commercial and industrial space, right? Work hard, get a pilot with a lighthouse customer delight them, scale out with them. use them as a reference, sign on brand-new customers and continue to build and 70% to 80% of our business kept coming from repeat customers who are very, very happy. .

That's exactly what we set out to do in AI, and we are doing this. Now with the utility scale customers, for the first time, I think they are seeing favorable regulation that allows them to rate base and offer better solutions to their customers. So we see a strong interest coming from both gas utilities and electric utilities to say in the face of that favorable regulatory and price design, rate design environment, can we partner with you.

We're always happy to partner with them. right? And our commercial industrial business is robust, strong and growing just the size of these big AI deals make us focus there. But trust me, we have a very active group prosecuting these orders. And think about the reshoring of big factories to America. How are they going to get the power -- we see that as a huge opportunity for us. So we are fully engaged in it. Our team is fully engaged in it. We just don't seem to talk more about it because of the size and scale of the AI opportunity right now. On the international front, look, it's very similar to what you're seeing everywhere else. -- on an 80-20 rule, pretty much the action today on AI and therefore, the huge power needs seem to be in the U.S., and that's what us and everybody else is focused on.

However, we will continue to develop them. We are continuing to develop them. And we believe that there will be a pause before it takes off very clearly what happened with Russia and Europe with natural gas followed up with what's now happening with Qatar and natural gas. Those things have an impact of slowing down development in those other countries. But we all know it's not if it will happen, when it will happen. So there is going to be a delay, and we are going to be prepared for it when that opportunity comes. Thank you.

And just my follow-up is on the cost front. With everything you have going on just with demand, could you just talk to us about your focus and kind of what you've done and what you plan to do on the stack life and just the total cost of the systems and kind of your pathway forward.

Ben, thank you. If you walk around the floors of gloom, there will be one thing anybody and everybody will tell you, gloom is about the genius of and. It's not about or, okay? So we don't accept false choices. It's not about growing demand only. It's not just about increasing capacity.

It's not just about reducing cost -- it's not about continuously not only about continuously innovating. It's about all of the above all the time. And people are sick of hearing me talk about the Genius of end. But that is really what's -- how we think or it's not in our recovery. So -- are we continuing to do that? Absolutely. Should you expect a double-digit cost reduction like we have over the last decade, answer is absolutely yes. We will still focus on that. And in terms of field performance, we continue to improve field performance, and that's the reason you're seeing our service margins to what they are. I'm very proud of our dedicated team. Thank you.

[Operator Instructions] our next question comes from the line of Colin Rusch with Oppenheimer.

Could you talk about the cadence of your installation times and how we should think about that trending into the balance of the year? And then just a little bit about the potential to leverage some of your available capacity into participation in project level economics for some of your customers?

Yes, that's a great question. So look, we have because we saw this [indiscernible] of demand coming our way and we understood that the primary driver there is going to be time to power. And we also understood that these large data centers prosecute on multiple projects. .

Just like any construction project, some are going to get delayed, some are going to be on do, some may speed up and they need that total flexibility. For all those reasons, we shifted from a have somebody big dirt for concrete, lay the conduits, get the trades people to come there and do all the work to a solution on a skid. That will just show up and get connected with the least amount of work that can happen. What is the result of that?

We have closed an order of magnitude reduction in the field time that it takes for us to be able to install our systems. That's a huge innovation. We have not talked about it at all. now that you're bringing it up, I'm just mentioning it to you. But that's again the genes of and. We just continue to innovate on every single area every single day.

So that's what we see. So I can assure you that we can get a 100-megawatt project up and running faster and with the least amount of field hours than any competing technology out there. So this is this is innovation and all. So thank you for asking that question.

Our next question comes from the line of Maheep Mandloi with Mizuho Securities.

And maybe just quick to first. First, just on the operating leverage. How should we think about that with the volumes to 5 gigawatts here? And separately on the service mix, how much of that 5 gigawatt would be for the service needs in the future here.

Thank you. I'm going to be very quick with that answer. The answer is very simple. We only talk about commercial product capacity. We always bake in our service requirements on its own that skip separately. So when we give you a number that is our commercial product revenue capacity.

And our final question comes from the line of Vikram Bagri with Citi.

Good evening, everyone, and welcome Simon to the team. We clearly are a better stock because that a lot of us on Wall Street. The first question I had here was you highlighted the culture of continuous innovation and improvement that leads to double-digit cost reduction which appears to be a significant advantage, as you highlighted throughout the call versus the competition.

I wanted to ask if there is price elasticity to demand. And I understand you benchmark our pricing against competition, it's apples and oranges, fully understand the benefits of the technology, whether it's speed to power, air quality, water backup and so forth.

We're still seeing premium being paid for CCGT, pricing being up 10% to 20% year-to-date this year. Is pricing something that if it goes down over time, we'll see more pronounced market share gains for Bloom from CCGT and the premium for CCGT getting eroded over time. better understanding of the product? Is it seeing the product work at a significantly larger scale at Toreo [indiscernible] of like leads to more market share gains I'm just trying to understand what's the tipping point where you see pronounced market share gains from CCGT and that premium sort of like getting eroded over time.

Thanks, Vikram. You're absolutely right on the cost reduction part. I'm going to indulge you and try to see if you would think about this differently. You're talking about some future tipping point.

The rate of our growth is faster than what any energy technology ever has done in the past, and that's what we see in our pathway. So this is -- for us, it's not a tipping point. Gone are the days , if you just go back to the traditional energy analysts from even a decade ago, just dedicated to 2015, 2016, the utility industries [indiscernible] Institute and all you analysts were talking about a downward spiral for electricity. This is not a zero-sum game. We don't care about what anybody else in the business does and who buys what from anybody else.

We are going to make sure that we are continuously improving our product to offer the best value to our customers, the best neighborly solution to our communities where they operate and create new demand and capture new demand because that new demand is going to be significantly larger than the industrial age demand.

The digital age demand is going to be significantly larger, and we are the digital solution to that digital demand. So we don't think about price. We don't think about cost. We don't think about elasticity. We think about meeting the needs and making sure that we win the AI race.

We don't -- or doesn't become an impediment to reassuring factories. Power doesn't become an impediment to electrification or doesn't become an impediment to digitization first in the United States and then use that model across the world, the power of the planet. That is really what this company is about, and I'm going to use your question as also my closing remarks since we are on the hour and simply state that what Hopefully, what you all understand is the following. Let me make a few statements and ask you to think about would you agree with it or not okay?

The use of AI and the amount of power that AI is going to use is going to go up and up over the next few years. The rate at which that growth has happened is not going to be met just by transmission and distribution upgrades. That means on-site power is absolutely essential. If on-site power is absolutely essential, in no neighborhood, would a community willingly want a power plant in their backyard that pollutes noisy and an is.

Bloom offers a no-compromise solution to both the digital customer and any community and neighborhood. If you bet against any one of the statements that I made you can bet against loan. Otherwise, you've got a great ride with us. Thank you for your attention.

And ladies and gentlemen, this concludes today's call, and we thank you for your participation. You may now disconnect.

Ladies and gentlemen, thank you for standing by. My name is Desiree, and I will be your conference operator today. At this time, I would like to welcome everyone to the Bloom Energy Fourth Quarter 2025 Earnings Conference Call. [Operator Instructions]

I would now like to turn the conference over to Michael Tierney, Vice President of Investor Relations. You may begin.

Thank you, and good afternoon, everybody. Thank you for joining us for Bloom Energy's Fourth Quarter and Full Year 2025 Earnings Call. To supplement this conference call, we furnished our fourth quarter and full year 2025 earnings press release with the SEC on Form 8-K and have posted along with supplemental financial information that we will reference throughout this call to our Investor Relations website.

During this conference call, both in our prepared remarks and in answers to your questions, we may make forward-looking statements that represent our expectations regarding future events and our future financial performance. These include statements about the company's business results, products, new markets, strategy, financial position, liquidity and full year outlook for 2026. These statements are predictions based upon our expectations, estimates and assumptions. However, as these statements deal with future events, they are subject to numerous known and unknown risks and uncertainties as discussed in detail in our documents filed with the SEC, including our most recently filed Forms 10-K and 10-Q. We assume no obligation to revise any forward-looking statements made on today's call.

During this call and in our fourth quarter and full year 2025 earnings press release, we refer to GAAP and non-GAAP financial measures. The non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between the GAAP and non-GAAP financial measures is included in our fourth quarter and full year 2025 earnings press release available on our Investor Relations website.

Joining me on the call today are K.R. Sridhar, Founder, Chairman and Chief Executive Officer; and Maciej Kurzymski, our Acting Principal Financial Officer and also our Principal Accounting Officer. KR will begin with an overview of our progress, and then Maciej will review financial highlights for the quarter. After our prepared remarks, we will have time to take your questions.

I now turn the call over to KR.

Good afternoon, and thank you for joining us today. Bloom is rapidly becoming the standard for on-site power as evidenced by our excellent fourth quarter, capping our best year yet. We delivered record revenue, gross margin and operating margin for the year. Our product backlog increased 140% year-over-year to about $6 billion. Our service business has been profitable for 8 quarters in a row. And in the fourth quarter, we achieved 20% gross margin in service with around $14 billion of service backlog and a growing product backlog that is 100% attached to service. Bloom is well positioned for durable growth in service revenue and profits in the years ahead.

Our growth has been fueled by seismic changes in customer attitudes towards power. Bring your own power has become the mantra for data centers and power hungry factories. On-site power has moved from being a decision of last resort to a vital business necessity. This shift has led large power users to seek Bloom to fulfill their needs. Our demand from data center and commercial and industrial or C&I customers is secular and growing.

In 2026, we will further invest in our commercial team to capitalize on growing sales opportunities. AI is a huge tailwind for the power industry and a big catalyst for Bloom's growth. The backlog we reported today includes half a dozen hyperscale and Neo cloud end customers compared to just one a year ago. Bloom has a master contract structure to enable these customers to keep returning to us for repeat orders, much as we have expanded with our C&I customers. And we are also experiencing surging demand in our C&I business. C&I backlog grew over 135% year-over-year, and it consists of several verticals, telecom, manufacturing, logistics, retail, health care and education. Digitization, automation, electrification and reshoring are driving C&I customers to seek on-site power. And our C&I sales pipeline is stronger than ever.

The geographic mix of our U.S. backlog is noteworthy. Two years ago, over 80% of our U.S. backlog was composed of installations in California and the Northeast, traditionally the high cost of power states. But this year, over 80% of our backlog comes from other states with lower power costs. This geographic shift highlights 2 important dynamics at play. First, companies are locating factories and data centers in states where they can quickly secure reliable and affordable power, either from the grid or on-site. The states where we are growing fastest have robust natural gas infrastructure and favorable regulatory and policy frameworks for on-site power generation.

Second, in these states with lower power costs, Bloom is cost competitive. Our value proposition, fast time to power, high reliability and lower emissions strongly resonates for our customers. In short, our customer base is diversified with numerous customers in every key sector, including AI. We are rapidly becoming the standard for on-site power. Given our healthy backlog and our robust funnel, I'm sure your questions will now shift from why we are expanding manufacturing capacity to when we will expand even more. Let me address that with some background.

At the core, Bloom is a technology innovator that rapidly delivers cost competitive platform products at meaningful scale to satisfy customers' current and future needs. We are building solid-state digital power for the digital age. We are not an industrial era energy company. Bloom's manufacturing IP and supply chain diversity enable us to scale without facing the multiyear delivery backlogs plaguing traditional suppliers. Our ability to scale also comes with a high ROI and low-risk profile. Capacity expansion requires a significantly lower upfront investment, a fraction of what legacy players need.

Our return on invested capital for capacity expansion is a few months, not years. This gives us the freedom to expand without predicting market size many years into the future to justify our deployment of capital. The simplicity of our manufacturing process is anything but simple. It represents years of innovation, thought and intellectual property. We have created a differentiated asset-light approach to manufacturing with the control and execution afforded only by in-house production and complemented with a diversified and global supply chain that flex us to meet market demand much like a tech supplier.

So my answer to questions on capacity expansion is simple. The Bloom Energy team reiterates its clear and simple promise to potential customers that have large time to power needs. Bloom will not be the bottleneck to your growth, and you can count on us to deliver timely power. We will deliver our power platform faster than you can build your greenfield facilities, be it an AI factory or a C&I facility. We demonstrated this recently by delivering a hyperscale AI factory order in 55 days against a 90-day commitment and power for a large factory before they could complete construction and commence operation. That is quick time to power, the Bloom way. In short, we will continue to expand deliberately and with discipline.

At a fraction of the cost and time it will take traditional legacy vendors. And we will offer our customers quickly deployable power that's reliable, clean and price competitive to meet their present and future needs. Speaking of future needs, let me address 800-volt DC. First, what is 800 volts DC? And why does it matter? The electric grid, turbines and engines were designed for the electricity loads of the 20th century factories and process industries. Large amounts of alternating current or AC power delivered at high voltage. 35,000 to 69,000 volts, contrast that to the needs of the digital age, computer chips, devices and other semiconductor equipment, everything digital in our modern world run on low voltage, direct current or DC power.

The upcoming AI computer racks will consume almost 100x more power than traditional CPU compute racks of Ester years to reduce copper use, increase efficiency and enhance compute density. AI racks will be architected to receive 800 volts DC. This switch to 800 volts DC is a necessity and not a choice and will happen at the compute rack level, irrespective of whether power is being supplied from an electric grid or on-site power. 800 volts DC will soon be the data center standard because physics requires it. Any AI data center using grid, turbines or engines will need to install numerous transformers, rectifiers and power conditioning tools to convert high-voltage AC to 800 volts DC. This adds significant cost, reduces reliability and increases emissions.

Bloom and only Bloom natively produces 800 volts DC today. No Band-Aids or adapters needed. Starting now, every Bloom server we ship will be 800 volts DC ready with a removable adapter that allows customers to deploy in legacy AC environments and migrate to DC on their own time line. This is a compelling future-proofed offering. We also offer to convert any servers we have shipped in the past to 800 volts DC with simple modifications, highlighting backward compatibility of this new future. 800 volts DC is one of our many innovative apps that integrate seamlessly on our energy platform, much like an app installed to a smartphone.

We will continue to make healthy investments in technology advancements this year and further strengthen our position as the innovative leader in the power sector. While we invest in the future, we'll continue to reduce costs of our core platform, keeping us on a path of anticipated margin accretion and further increasing our advantage over traditional solutions. We look forward to a strong 2026 as we continue our journey to become the standard for on-site power, a benchmark for speed, reliability and customer value in the digital age.

Over to Maciej now for a financial overview. I'll join you in a few minutes to answer questions. Maciej?

Thank you, KR, and good afternoon, everyone. On today's call, I will discuss results of both the fourth quarter and the full year and also provide our full year 2026 guidance. Let me start by recognizing all of our employees at Bloom for incredible execution in 2025 by calling out 3 highlights that the team drove this year.

First, we achieved record financial results in several key metrics. I would like to highlight the $271.6 million in adjusted EBITDA, proving just how much operating leverage there is in the business as we start to scale. Second, we were free cash flow positive for the second consecutive year. And third, our service business achieved approximately 20% non-GAAP gross margin for the first time. None of that would be possible without the fantastic performance and dedication of the entire Bloom team.

As a reminder, I will focus my discussion on non-GAAP adjusted financial metrics. For a reconciliation of GAAP to non-GAAP, please see our press release and the supplemental deck on our website. Revenue for the quarter was $777.7 million, up 35.9% year-over-year. On-site power continues to accelerate relative to the grid and Bloom's ability to deploy our energy servers and power up sites in record time continue to highlight Bloom's value proposition and drive revenue growth.

Gross margin was 31.9%, lower than the 39.3% gross margin in Q4 of 2024. Gross margin will continue to fluctuate given the mix of individual projects in the quarter, but we will continue to manage this movement through product cost reduction efforts and operating expense efficiencies, leading to a stronger EBITDA. Our operating income was $133 million versus $133.4 million in Q4 last year.

Adjusted EBITDA was $146.1 million versus $147.3 million in Q4 2024, while EPS was $0.45 versus $0.43 a year ago. Again, these are all non-GAAP results. Our product margins were 37%, while our service margins were approximately 20%. This is the third straight quarter of double-digit margins in the service business. And while we will see some volatility in these results on a quarterly basis, we expect to continue to see annual improvement.

Our balance sheet is much stronger than a year ago as we added significant cash through convertible bonds. We ended the quarter with $2.5 billion in total cash on the balance sheet. Our inventory ended the year at $643 million, slightly higher than what we expected at the beginning of 2025 as we prepare for a strong 2026. Our cash flow from operating activities was an inflow of $113.9 million, while CapEx was $57 million.

Turning to the full year. Revenue was a record $2 billion, up 37.3% from 2024. Non-GAAP gross margin of 30.3% was up from 28.7% in 2024. Non-GAAP operating profit of $221 million, up $113.4 million from the previous year on a revenue increase of $550.1 million or 20.6% drop-through to operating income. Non-GAAP gross profit in our service business was $29.7 million, a significant improvement from 2024, as I mentioned earlier. Service was profitable on a non-GAAP basis during every quarter of 2025 for the second consecutive year. Looking forward, we continue to expect to drive improvements in service profitability as we expand our installed base and scale.

Before we get to guidance, I want to talk a bit about our backlog. We see tremendous momentum in commercial engagement across both the data center market as well as C&I. Our product backlog has more than doubled from a year ago. We also have approximately $14 billion in service backlog. We have grown our backlog while maintaining healthy customer mix and do not have oversized concentration to any one customer. This brings us to guidance for 2026.

While 2025 was a great year for Bloom, we expect 2026 to accelerate. We expect 2026 revenue to be $3.1 billion to $3.3 billion. Non-GAAP gross margin of approximately 32% and non-GAAP operating income of approximately $425 million to $475 million. We expect capital spending to be $150 million to $200 million and cash flow from operations to be close to $200 million. We do expect to invest in our R&D road map and commercial efforts. But as you can see from our operating income projections, we expect to capitalize on the significant operating leverage of the business to drive profit expansion.

To conclude, Bloom's disciplined execution is delivering accelerated growth while maintaining sustainable profitability as we scale. We believe that we are rapidly becoming the standard for the on-site power generation market, and I could not be more excited about the opportunity in front of us. Operator, we are now happy to take questions.

[Operator Instructions] And our first question comes from the line of David Arcaro with Morgan Stanley.

I was wondering, could you speak to the follow-on opportunities at existing customers? I was curious, how have the initial projects gone? And how seriously are some of those customers now considering follow-on orders with you?

David, nice to hear from you. Yes. Look, that's a very important question you're asking. You build a strong company on the basis of happy customers. We have seen that from the day we started. We can tell you even in our commercial and industrial business that has been our traditional bread and butter, over 2/3 of our business year-over-year comes from repeat customers bringing in multiple repeat orders to us. This is how it operates once people get used to Bloom, they love Bloom because we keep our promise and we deliver to them.

So this is no different right now with the newer sectors. And the customers that have lately engaged with us, be it our utility partners, be it a hyperscale end customer is seeking us out once they have tried us out and that's strong traction. Oracle would be a very good example that is -- that's happening on a daily basis. We have great conversations with them on so many projects, and we are working with them on many projects -- many prospective projects in the future.

Okay. Excellent. I appreciate that color. And also some of your thoughts on further manufacturing capacity expansion. I was wondering, as you consider the potential for the next one, the next capacity expansion, just curious what are the milestones or triggers that you'd be watching for? And when do you think a decision could potentially be made about that next tranche of additional production capacity?

Look, for us, making a decision on expanding is like everyday business for us. It's not a major issue for a very simple reason. We are extremely capital light in order for us to expand. We have standing orders with our suppliers of equipment, with our supply chain, with everybody else to ramp up as quickly as we need to. And the return on investment for us is a few months. So when do we make a decision? It's fairly simple. If we see a large opportunity on a time to power and we need to be able to expand our capacity to be able to provide that additional power to a customer, we can ramp up and provide that additional power to that customer before they are ready.

Typically, it takes more than a year to stand up a greenfield data center. It takes more than a year to stand up a factory from permits all the way to full implementation. We can be ready for them before then. So this is a continuous decision we will make going forward, quarter after quarter. The reason we signaled to you last year that we are going from 1 to 2 gigawatts was, there was concern in the market about do we have a pipeline, do we have an order? We just wanted you to show how much confidence we had. So we signaled that. And now we all understand why we are expanding. But going forward, we'll just continuously keep expanding our capacity, and that's just normal business for us.

Our next question comes from the line of Christopher Dendrinos with RBC Capital Markets.

Congratulations on the strong quarter and year-end outlook. I guess to start here, I wanted to follow up on the HVDC architecture. And I guess, when do you think we could start seeing that solution set deployed? And how are those conversations going with customers? It seems like you all will have quite a bit of an advantage here from a cost perspective without needing to deploy all that extra electrical equipment. So I'm just curious how that's shaping up in the pipeline and how you're thinking about that opportunity.

Look, I think -- thank you for that question. And thank you, Chris, for your sentiments about the quarter and the year. We are super excited about the accelerating momentum that we are seeing in our business as we sit here today. And clearly, we just, I think, add even more of a competitive advantage by bringing in the 800-volt DC. And I think you're all beginning to understand why that's important from whether it's a CapEx perspective, a reliability perspective, use of copper and lack of availability of copper and transformers with the alternative option as well as the operating cost because of efficiency losses you get from -- switching from extremely high-voltage DC or medium voltage DC to the 800-volt DC, all those reasons you understand.

Now we are betting like we bet 25 years ago, DC architecture is the right way to go, and you can see where we are today. And we are betting that very quickly this solution is going to be sought after. And anybody who's not implementing that on day 1 for their data centers now because they don't have the supply chain on their sites ready for implementing that architecture will want to switch our equipment to that DC as soon as they are ready. That's the reason we're going to ship everything going forward as 800-volt DC.

Your question on when will the data centers be ready for it, that's better left to ask them as opposed to us. We are always going to future-proof them and be a step ahead. The key to being a great supplier in the digital economy is you're anticipating what their needs are and you're there ahead of when they need it. And that brings everyone forward with greater speed. We are moving at AI speed on this one. Thank you.

And I guess maybe as a follow-up, just sticking on that point of anticipating customer needs. I guess on the R&D side of things and the technology road map, what's the next kind of step for Bloom here? And how are you thinking about the evolution of the product?

Look, we are working on a lot of apps right now. But I'll tell you one that I think we have talked about a little bit here and there in the past. And one I'm super excited about and makes our customers and potential customers who come out to take a look at our operating systems in the lab is this rapid load following of the AI load being handled by our systems without requiring batteries. That is huge. And the ability to, in an islanded mode, operate our systems without needing any backup in terms of backup generators because of our high reliability. But add to that, not needing batteries to keep up with the wild swings that the AI load have in terms of power is a super important application.

And every day, we are making that better and more robust. And any time a potential customer throws an AI load profile at us, our team is able to just seamlessly integrate that into our product and show them why it will work really well. That's a huge advantage, not just in terms of cost savings, in terms of safety of avoiding all those batteries inside a data center complex and the -- like fire assets that -- like that may create the maintenance issues it may create. On top of that, think about this, as the AI data centers grow, that battery supply chain also becomes a constraint, and we are completely eliminating that constraint, not even mitigating it.

So super excited about applications like that. There'll be many more to come as the time goes by. Stay tuned. And all that I can tell you is we have a lot more ideas of a lot more apps that are going to reside on the smart platform.

Next question comes from the line of Manav Gupta with UBS.

Congrats on a very strong quarter, KR. I wanted to ask you about the progress you're making on combined heat and power solutions. Most data centers are using vapor compression chillers powered by electricity. I think your absorption chillers would be using thermal process that is powered from waste heat. At this point, vapor compression chillers are the primary source of cooling and then absorption chillers are being deployed for some supplemental cooling. I'm trying to understand in a world where electricity is very expensive and grid power is not available, can absorption chillers actually grow at a faster pace than vapor chillers? And if it does happen, then how does BE benefit? Your product already has an 800-volt advantage. Do absorption chillers make it even more competitive?

Manav, I wouldn't have expected anything other than a strong technical question from you. Like kudos to you on your very good research report on the 800 volts DC, I think it's a must-read for people to understand that. Well done on that. So on the absorption chillers, here's the answer, right? Thus far, wafer compression was being used simply because the energy coming into a data center came in the form of electricity from a wire. The generation facility that made that electricity was made hundreds of miles away somewhere, and therefore, you couldn't pipe the heat, the excess heat all the way from that far away generation capacity to where the data center is.

Now with on-site power generation being the go-to option, a necessity option, for data center customers. If we are generating power for them on site, in addition to our extremely high electrical efficiency, we have high-quality heat, and that heat is allowed to drive a very well-established technology called absorption chilling to provide cooling. We think we can reduce electricity usage in the data center by at least 20% to 0%. That's a big number for this huge power-hungry gigawatt plus data centers. And what do we do with that? It's chilled water at somewhere around 5 degrees Celsius or 40-degree Fahrenheit coming in. We have systems now that we are operating in this mode chilling and cooling our factory just to demonstrate to customers. Customers are super interested in the solution right now, a, because it is more efficient, less expensive. And there's an additional environmental benefit coming out of those absorption chillers in that they don't use hydrofluorocarbons, and that is a big issue for global warming.

And lastly, by using absorption chillers where they do on-site power generation, they are not competing with the same supply chain constraints that they have to on the vapor compression. For all those reasons, this is extremely important. So think of this as another app on our platform and you asked the question of, does it make us more competitive? A smartphone that has more apps and can solve more problems for a customer is always a more competitive solution. That is what we are quickly becoming out here, as you can see.

Congrats on all the positive developments that are happening in your company, sir. Congratulations. I'll turn it over.

Next question comes from the line of Davis Sunderland with Baird.

Please let me echo the ones before me and saying congratulations on a great quarter and a great year. I wanted to ask, your guys' technology is increasingly being comped to legacy incumbents such as combined cycle gas turbines. Could you maybe talk about if you guys are seeing project wins against these or other types of technologies?

So our answer to you, I won't discuss about the competitive landscape. That is something you should ask the end user in terms of what did they compare before they choose us. But you can clearly see from the stamp sizes of the projects we operate, we are operating in that class, very clearly. And we are no longer the 10- and 20-megawatt systems. We have hundreds of megawatts going into the gigawatts very soon kind of single site location, that is the stamp size you're looking at.

So very clearly, it's in the same category of a combined cycle gas turbine, right? If you consider the entire value proposition, not just in any one narrow aspect from a customer's point of view, if it is on-site power that is islanded, can a combined cycle gas turbine operate and provide power at partial loads? Can it swing up and down with the needs of the load? Can it -- what happens to that system in high altitude? What happens to their efficiency? Can you modularly pay as you grow and operate that system? Or is it monolithic, okay? None of the features of a very large stamp size, anything really matches with how a digital world of a data center operates, okay? When you've taken the cost associated with all that stuff and then very large mechanical equipment with its inertia, cannot swing up and down in milliseconds and seconds like our solid-state digital platform does.

So it cannot follow a load that way. So you need Band-Aids for that. And then a huge combined cycle gas turbine because of the large amount of power it puts out monolithically can only do that with reasonable amount of copper at very high voltage. And you need Band-Aids to now be able to bring that to 800 volts. You put all that together, can we compete? Yes, yes, we can. And the fact that we are winning these kind of stamp sizes should show you that we are able to do that not just in high-cost value places, but in states where cost of electricity is traditionally low.

So even in a place where irrespective of gas prices, irrespective of utility prices, we are able to compete. We're soon becoming the standard. And the customers who evaluate the entire value proposition will choose us. Customers who just look at first cost, surely, if you can find it and if you can install it in a short amount of time, you'll be able to use a combined cycle gas turbine.

And as a follow-up, could you talk just a bit about how the life of fuel cell stacks has improved, maybe how it relates to service margins and how you guys think about risk in the services business?

Yes. So thank you for asking me to highlight that. That should no longer be a question on any one of your minds. We fully understand when we were losing money every single quarter on service, and we told you that we have a road map to get to gross margin neutral and then gross margin positive and keep accreting. You had a reason to wonder about tell us the specifics, show us that you're making progress. What you're looking from the last 8 quarters is 8 continuous quarters, contiguous and continuous quarters of profits in the service business.

And on top of that, Q4 of 2025, we had a 20% gross margin. And as our fleet sizes increase, as our technology keeps getting better, and if you look at our $14 billion backlog in service, and then you understand that every order that we are booking has a 100% attach rate to service and will add to that backlog. You will clearly see that service is going to be a growing profit generating, revenue-generating business for Bloom for years to come, which is going to be a huge advantage. This is the reason why we have worked so hard on it, and we'll continue to work hard on it. Improving life, reducing cost, operating the system with AI-driven digital platform.

Let me highlight that for a second. Here is what you need to know. We have a few trillion cell hours of field operation is what Bloom has, a few trillion cell hours. More than 6 billion data points come from our field to us every single day. We are using AI. We are not only benefiting from AI on our revenue side. We are using AI to our benefit for all this to improve our performance every single day because we have a digital twin associated to every single fuel cell stack and data from the real field is coming and feeding the digital twin and making our models better and better.

So this is how we are going to build that business. It's a strong business for us. You shouldn't have to worry going forward about what are we going to do with service business. The more important question is, are you placing enough enterprise value to this service business?

Next question comes from the line of Michael Blum with Wells Fargo.

Congrats on the quarter. So I'm wondering if you could speak to one of your suppliers, MTAR Technologies had extremely bullish comments on the earnings call, projecting 30% growth CAGR to 2030 for Bloom Energy. So I'm wondering if you could just speak to that and maybe help us square that with the updated backlog number.

We are appreciative of the enthusiasm that our strong supply chain partners have and how bullish they are about what we do. But Michael, either to all of you or to our Board or to our vendors, we have not provided any long-term guidance. And you can't attribute any of that to us. You would have to ask them where their confidence is coming from and square that with your own models. But we don't -- we have not provided any guidance that far out.

And again, this is not a flip in comment. Let me tell you what's just happening, right? Just take the last 3 days, last 2 days of what you're all seeing in the market. Amazon came out today along with us after the market and said they're upping their capital expense almost 100% to $200 billion for the year 2026. right? Google did the same thing yesterday or -- yes, yes, yesterday after the bell. And again, upping their CapEx heavily to $175 billion to $185 billion. This is all for the digital infrastructure. You know what, what you're seeing happening is the horizon at best is 6 months, long-term horizon. Nobody has visibility past that because this entire field is accelerating at that pace. For us to sit here and talk about 2030, that's the old industrial age resource planning that the utility companies used to do. That's not where the digital age is going. We don't have any predictions for 2030 right now other than to say we are extremely bullish and it's going to accelerate. Thank you.

I appreciate the clarity there. Other question I had was on the backlog. I'm wondering if you could tell us what the mix is, U.S., international. And really, the broader question is, if you could speak to your conversations you're having with prospective customers. Should we expect most of your business is going to be in the U.S. going forward? Or is there a meaningful international market opportunity also that we should be thinking about?

Thank you. That's a good question. We don't break down the mixes between U.S. and international. But look, to answer your broader question, Bloom is going to be a global company. We are going to expand and really play a major role in other countries. That is going to -- actually, if you think about what is the kind of infrastructure we need to be able to play in those areas, it's going to lag behind the U.S. simply because LNG terminals, the amount of LNG available for new projects, given what's happened in the world with like Russian gas being cut off to Europe, things like that, is necessarily going to take a few more years to take off in a big way. If there is gas going to these countries now, it is to support existing infrastructure. It's barely available to support new growing infrastructure. And you're not seeing very large projects in Europe for that reason. You're not hearing about the 0.5 gigawatt and gigawatt data centers being built out there. right, other than where power is already available.

So it's going to lag behind a little bit, but we are going to stay on top of it. A predominance of the opportunities right now for everybody in the world is here in the U.S. The growth rate is unbelievable. So do we see at least for the foreseeable future, this being the key area -- U.S. being the key area of focus? The answer is yes. In terms of the diversity of the mix, I want to remind you all, as much as we talk about AI, commercial and industrial business is very strong for us. We have had a 135% growth year-over-year in our commercial and industrial backlog. Companies, factories, campuses, retail businesses; they're all digitizing. They're all automating. They're all using robots. They're all seeking AI, their power needs, their power draw is going up.

So that electricity demand is very high. And when a factory is getting built, they can't wait for the power company to give them the power at their own pace. So there -- we see them coming more and more to us. And it's all happening in the middle of the country where there is gas availability and where there's proper policy for on-site power being encouraged.

Next question comes from the line of Colin Rusch with Oppenheimer.

Guys, as you look at the depth of the market and the breadth of customers that you're dealing with and the value-added elements that you have to your system with the future-proofing and cooling dynamics, can you just talk about your pricing strategy? As you get a little bit deeper into this, how much pricing leverage do you have? And how much do you want to take here over the next 12 to 24 months?

Look, pricing really is very much a market phenomena like based on where people are. People are now going to states where they can get affordable power. But affordable power, remember, is value-based, okay? Most of our customers place value on time to power, place value on ease of permitting because we don't create air pollution. And they don't want to get caught in a backlash or either a non-permit or a backlash from their local community.

And we feel very, very good that our customers truly value the proposition we bring to them. And so if you just looked at where we are, we don't see us having to choose between growth and profitability, okay? -- between our continuous cost reductions and efficiencies and given where electricity prices are going elsewhere, and if you just listen to the legacy suppliers of turbines and engines, they're all talking about pricing leverage. What does that mean? They're actually increasing their prices.

So electricity for customers is going in only one direction. What we offer is really a competitive price, but at a value stack that they're extremely happy with and are willing to pay. So I don't think in the foreseeable future, we have to be looking at worrying about pricing.

And then the follow-up here is really around any interest in potential M&A. Obviously, you've got a lot of wood to chop with the core product here. But with an augmented balance sheet and a very robust currency right now with the stock, is there any reason or opportunities for you guys to start looking at incremental acquisitions to scale the platform at all?

Look, we can be selective about things that matter to us and things that matter to our customers. And if we had some acquisitions, will that make it easier for us to bring that entire smart platform to our customers in a better way. Other than that, our potential addressable market and our ability to light up the planet is just unthinkably big that we don't need to be looking at what else should we be doing. If we just -- like lighting up the planet is a good day job, I don't need another day job.

Next question comes from the line of Mark Strouse with JPMorgan.

KR, I thought it was really interesting when you said that over 80% of the backlog today is in some of those lower-cost states outside of California and the Northeast. Appreciating maybe some of that's driven by data centers. I was curious if you could maybe give that metric for your non-AI business, kind of what that mix might look like?

No. Sorry, we just don't do that. Give me another question, I can answer you.

Okay. All right. I'll follow up offline. Can I ask on the book and ship business. This time last year, I think you said 2024 was the first year that the majority of your revenue came from book and ship. Can you talk about what that looked like in '25 and how you're thinking about that going forward?

Yes. Yes, sure. And that's an important thing, right? We in like '25, we had a significant double-digit percentage, let's just put it that way, of book and ship that we were able to do, booking, shipping, turning power on for our customer. And you heard one example of that where we powered a data center like 55 days, right? So -- and so that was a significant part of the business. And we would expect there are plenty of our valued customers who are going to come to us and want that power very urgently for whatever reason they have.

And most often, I can tell you, Mark, it is some other vendor who did not keep their promise and they come to us. We see this as a competitive advantage. And we want to be able to support a customer under those circumstances. So we have the capacity to do that. We would love to do that. And I would think it will still be in like double digits in terms of percentages.

The next question comes from the line of Sherif Elmaghrabi with BTIG.

So last month, AEP exercised their option for fuel cells under that gigawatt agreement with Bloom, but the offtake won't be finalized until the second quarter of this year. So my question is, would you expect them to take delivery of the fuel cells regardless given the existing power demand environment and the infrastructure they put in the ground as well?

Yes. AEP very clearly said that in their press release, and we reiterated that, that our sale of that product is unconditional, and they will take possession. And obviously, they wouldn't have accepted that if they didn't believe very strongly that they could get this going, number one, in terms of that project.

But additionally, here is what you need to know, right? AEP and us are working on several projects together, and they're great partners of ours. And we expect that our combined business is only going to grow. So they didn't have any concerns about signing a definitive order with us, even though they had to go through some formalities on their side because Bloom's energy servers are not a perishable. They can easily put that to use in multiple other locations that they are potentially considering us for.

Second, I do want to ask about the warrant transaction with Oracle. That deal helps align your interest, of course. And I'm wondering how do you think about doing similar transactions with other hyperscalers if that's something they're interested in?

Again, we still haven't executed that agreement. As you know, we are working through that strategic partnership agreement that we have. And because of that, I can't speak to the details of it because it's not out there. But you'll see that soon. Everything is on a case-by-case basis. In this particular case, I'll tell you what the criteria was. It was a great strategic partnership where both enterprises had lot to gain. And by doing that -- and remember, these are not penny warrants. These were done at market pricing on the day we agreed to like what we do. So it is not in lieu of something other than both parties enhancing enterprise value. So if -- so I'm not going to say yes or no to this. It will all be evaluated on a case-by-case basis if there's enterprise value. So it's -- so the answer is neither a yes or a no. It depends.

And our last question comes from the line of Noel Parks with Tuohy Brothers.

I wanted to ask about the product margins and the supply chain. I was just wondering what your thoughts are on your visibility into your input costs for components. And I'm just wondering if there was any trend you're considering towards longer-term contracting or forward purchasing from vendors to sort of exercise your leverage with cost.

No, thank you so much. And I'm going to ask for your indulgence and say, let me not do a follow-up question because we are like running on the hour, but let me answer this very good question you asked. Look, we are constantly working both internally in the company and with our supply chain partners to figure out efficiencies, how to bring scale-related efficiencies, how to bring about technology and process-related efficiencies and continuously keep bringing down the cost.

We are also extremely judicious of watching what is going on in the world and securing at risk buys if we need to, if we see certain things happen. You can see we are not pinching the last penny on the amount of inventory we hold. There are very good reasons for those things because we're very strategic about all those decisions. And in terms of cost reduction, overall, double-digit cost reduction is in our DNA every single year, and we make that happen. Sometimes all of that translates out. Sometimes because suddenly a tariff regime came along instead of saying our cost went up, we're able to neutralize a lot of that or during COVID, when the cost of logistics went up, we didn't have to increase our price because we could make up for that cost increase using our cost reductions. But we have always delivered that, and we'll continue to deliver that, and that's in our DNA.

So that's how we are going to bring about margin accretion over a long period of time in this company and keep growing our margins, okay? With that, let me conclude to say in the closing, look, you all see Blooms really executing from a position of strength, but we are also scaling with discipline. And we are on a very firm path to make sure that we become the standard for on-site power, the benchmark for speed, reliability, flexibility, everything. As you heard in the last couple of days from the large digital companies, they're all increasing their CapEx by amounts that would have seemed unbelievable even 2 years ago. These numbers are staggering.

And this is all for CapEx infrastructure. Everything is digital. This is digital infrastructure. Digital runs on electricity. Electricity at that pace cannot be delivered by anyone in the free world today using poles and wires. On-site power is a necessity. Bloom brings very clear competitive advantages that legacy providers that built their technology for the industrial age cannot adapt to. So we are very confident in the path we have charted for ourselves and are excited for the future. We look forward to another very good year. Thank you so much.

Ladies and gentlemen, that concludes today's call. Thank you all for joining in. You may now disconnect.

Ladies and gentlemen, thank you for standing by. My name is Colby and I'll be your conference operator today. At this time, I'd like to welcome you to the Bloom Energy Third Quarter 2025 Earnings Results. [Operator Instructions]

I'd like to turn the call over to your host today to Michael Tierney, Vice President, Investor Relations. Sir, you may begin.

Thank you, and good afternoon, everybody. Thank you for joining us for Bloom Energy's Third Quarter 2025 Earnings Call. To supplement this conference call, we furnished our third quarter 2025 earnings press release with the SEC on Form 8-K and have posted it along with supplemental financial information that we will reference throughout this call to our Investor Relations website.

During this conference call, both in our prepared remarks and in answers to your questions, we may make forward-looking statements that represent our expectations regarding future events and our future financial performance. These include statements about the company's business results, products, new markets, strategy, financial position, liquidity and full year outlook for 2025 or 2026. These statements are predictions based upon our expectations, estimates and assumptions. However, as these statements deal with future events, they are subject to numerous known and unknown risks and uncertainties as discussed in detail in our documents filed with the SEC, including our most recently filed Forms 10-K and 10-Q. We assume no obligation to revise any forward-looking statements made on today's call.

During this call and in our third quarter 2025 earnings press release, we refer to GAAP and non-GAAP financial measures. The non-GAAP financial measures are not prepared in accordance with U.S. generally accepted accounting principles and are in addition to and not a substitute for or superior to measures of financial performance prepared in accordance with GAAP. A reconciliation between the GAAP and non-GAAP financial measures is included in our third quarter 2025 earnings press release available on our Investor Relations website.

Joining me on the call today are K.R. Sridhar, Founder, Chairman and Chief Executive Officer; and Maciej Kurzymski, our Acting Principal Financial Officer. K.R. will begin with an overview of our progress, and then Maciej will review financial highlights for the quarter. After our prepared remarks, we will have time to take your questions.

I will now turn the call over to K.R.

Good afternoon, and thank you for joining us today. I'm delighted that Bloom had its fourth consecutive quarter of record revenue. This seminal year for Bloom positions us for an even stronger 2026 and beyond with higher growth and more profitability. Three major tailwinds benefiting Bloom today have created a once-in-a-generation opportunity for us to become the global standard for on-site power generation.

First, the AI build-outs and their power demands are making on-site power generated by natural gas in necessity. Second, winning the AI race is a nation state priority, driving government policy and removing barriers that had previously been headwinds for on-site power generation. Third, our product innovation is advancing at a pace more akin to semiconductor evolution than to that of traditional industrial products.

Every year, for over a decade, our fuel cells have seen double-digit year-over-year cost reduction. While our costs are coming down, our performance is going up our fuel cells last longer, are more reliable and are more efficient and today produce 10x more power in the same footprint than they did 10 years ago. These improvements have opened up large market opportunities.

For example, we historically sold exclusively in high-cost electricity markets such as California and the Northeast. We are now competitive in large power hungry markets of the Midwest, Mid-Atlantic, Mountain West and Texas and many European and Asian cities. Bloom is now positioned to become the standard in on-site power, which many of us believe will be a $1 trillion market.

Becoming the standard means we will be the benchmark by which all others are measured. The reference point for speed, reliability and performance in on-site power and customers, partners, regulators and governments think about dependable dispatchable electricity. They should think about Bloom first. While we built Bloom with the conviction that this moment would arrive, we had no illusions of the difficulties we would face to gain acceptance as we embarked on this journey.

To be even considered, we had to be better in every dimension. We persevered and delivered step by step. Now after 24 years, we have robust supply chains, manufacturing processes, installation capabilities and field performance data to show our customers we offer an unparalleled on-site power solution at speed and scale. We obsess about meeting our customers' needs and do not expect them to compromise. We do not offer them false choices, clean, reliable or fast instead we offer them and solution. We ship on time and aim to ship faster than anyone else. We are more reliable and resilient and offer our customers superior price-to-performance value.

Our mass-produced modular power systems allow us to power sites as small as your neighborhood retail store and as large as a Giga AI factory that mass manufactures intelligence. Bloom Energy Servers are safe, operate without consuming water, do not pollute the local air and have curb appeal. All features that make them welcomed in the communities where they are installed.

The precursor to becoming the standard is to first earn our place in the evaluation process alongside the well entrenched and very capable competitors that have defined the market for decades. We are now executing on this phase, working to replicate in new markets, the success we have achieved in sectors like semiconductor manufacturing and telecommunications, industries that demand the highest reliability. Today, we are the standard for on-site power in telecom and semiconductor manufacturing as evidenced by the rapid adoption of our technology by the top-tier players and the strong sales pipeline in those segments. Our strategy is deliberate and simple, in each vertical, we establish our credibility with a lighthouse account and then build on that success with other Tier 1 customers.

For example, in telecommunications, we first secured AT&T as a lighthouse customer in 2011. After they became convinced of our operational excellence, they deployed us in multiple sites in many states. Soon, we added Verizon and T-Mobile as customers and have sold over 100 megawatts of on-site power to telecoms. Today, we are a go-to on-site power choice for U.S. telecom companies.

Now we are following the same playbook to become the standard on-site power solution for AI. We are embedded in 7 distinct AI ecosystem channels. In each channel, we have secured a lighthouse customer in our robust pipelines.

First, the hyperscalers. Back in August, we announced our first deal to power an AI factory with Oracle. We have fulfilled our delivery ahead of schedule. We promised to deliver in 90 days, and we delivered in 55 days.

Second, electricity providers. Last year, we signed a gigawatt agreement with AEP, which purchased our fuel cell systems to power another big hyperscaler, AWS.

Third, gas providers. We signed our first deal with a major gas provider who will convert its gas to electricity with Bloom fuel cells and sell that on-site power to a third hyperscaler. The hyperscaler will announce details of this installation when it is ready.

Fourth, our co-location providers. We work with many, including Equinix, which has deployed over 100 megawatts across data centers in multiple states.

Fifth, Neo-clouds. Our systems are generating on-site power for top new cloud provider, CoreWeave, at a high-performance data center in Illinois.

Sixth, data center developers. When an understanding has been reached on key terms, developers begin to file permits and permissions. You may have seen some of these public filings recently.

Seventh, infrastructure owners. Large infrastructure funds are increasingly developing their own AI factories. Brookfield, the world's largest AI infrastructure investor has invested $50 billion in AI opportunities and is tripling the size of its AI strategy over the next three years. It announced an AI infrastructure partnership with Bloom Energy and made an initial investment of $5 billion. Bloom will be the preferred on-site provider for Brookfield's trillion-dollar infrastructure portfolio of AI factories, data center operators, corporate facilities and factories. Brookfield will also finance Bloom-sourced AI opportunities. We have already completed projects and Brookfield plans to announce a Bloom-powered European AI inference data center project by the end of the year.

To recap, we have strong traction across all channels of the AI ecosystem. Each channel is anchored by a lighthouse customer and accompanied by robust commercial activity. As we continue to penetrate new geographies and verticals, success builds upon itself and should make each new market entry easier than the first. The opportunity is vast, and we are still in the early innings.

So what are we doing to make sure we are ready to handle growth as well as further advance our leadership position? As we have previously announced, we are doubling our capacity to 2 gigawatts by December 2026, which will support about 4x our 2025 revenue. That expansion is all systems go. Bloom's capacity will not be a bottleneck for our customers. We are also investing in operational talent and capabilities needed for the expansion of our production capacity beyond the 2 gigawatts. We are building a commercial team that can capture opportunities across diverse market segments and geographies. And we are continuing to invest in R&D to increase our lead in on-site power. We are doing all of this while maintaining our focus on operational excellence and financial discipline to achieve margin expansion over time.

Based on what we see today, we expect 2025 to be better than our previously stated annual guidance on our financial metrics. In addition, we expect double-digit product cost reductions to continue and keep us on a path of margin accretion. We look forward to a strong 2026 as we march forward and build a future where Bloom powers the digital age and is the recognized standard for on-site power globally.

I'll turn over to Maciej now, and I look forward to answering your questions.

Thank you, K.R., and good afternoon, everyone. As K.R. mentioned, Bloom is now positioned to become a standard in on-site power, our announced customer base and financial results are a testament to this. On today's call, I will discuss our Q3 financial performance and make a few comments about fiscal 2025.

The last four quarters have been a record operational and financial performance and Q3 was no exception. While our commercial success have been most visible, the work our engineering, manufacturing and support teams have done [indiscernible] to drive product cost reduction is evident in our financial results. Highlights include record third quarter revenue, positive cash flows from operating activities and our seventh consecutive quarter of profitability in our service business.

As a reminder, I will focus my discussion on non-GAAP adjusted financial metrics. For a reconciliation of GAAP to non-GAAP, please see our press release and the supplemental deck on our website.

Revenue for the quarter was $519 million, up 57% year-over-year. Time to power needs are creating demand for on-site power. This, together with the advantages of our fuel cell technology for AI factories is driving our revenue growth. Gross margin was 30.4%, 510 basis points higher than the 25.2% gross margin in Q3 of 2024, driven by continued focus on product costs and manufacturing efficiencies, our operating income was $46.2 million versus $8.1 million in Q3 last year. Adjusted EBITDA was $59 million versus $21 million in Q3 of 2024 while EPS was a positive $0.15 versus $0.01 loss a year ago. Again, these are all non-GAAP results. Our product margins were 35.9%, while our service margins were 14.4%. This is the second straight quarter of double-digit margins in the service business, and we expect this trend to continue.

As we have talked about on each call this year, we took advantage of our balance sheet and a visibility into customer demand to level load our factory. We expect to work down inventory in Q4 as our shipments of product accelerate. Cash flow from operating activities was an inflow of $20 million, primarily due to working capital improvements. We ended the quarter with $627 million in total cash on the balance sheet.

Turning to the full year. As K.R. mentioned, based on what we see today, we expect fiscal 2025 to be better than our previously stated annual guidance on our financial metrics.

To conclude, Bloom is focused on not just on scale, but on showing sustainable profitability as we grow. We are uniquely positioned to benefit from this unprecedented market dynamic, and I could not be more excited about the opportunity in progress.

Operator, we are now happy to take questions.

[Operator Instructions] First question comes from the line of David Arcaro from Morgan Stanley.

Great. I was wondering if -- a very helpful commentary, too. I was wondering if you could talk about the pace of commercial activity that you're seeing. You've had success now with multiple agreements in a short period of time. How do you see this playing out as we look forward to the next agreements in the pipeline just in the context of the market demand that you're seeing?

David, thanks for that. Look, I think we've been saying in the last three earnings calls that the commercial momentum is robust. And all that I can tell you if I looked at it this week and last week, and if I walk over to the commercial section of our offices, is that momentum is clearly accelerating, and it's palpable. Okay. So forget questions of, is it static or is it slowing down? It's accelerating. That's all we see. And we see that across the board. And by the way, it's accelerating not just in AI, our traditional, commercial, industrial segments are doing the same. So it's across the board. And the larger the deals get, the more the actors that get involved, as I explained in the entire AI value chain. These are complex deals. And each one goes to a different phase, different momentum. Some close extremely fast because [indiscernible] take a little bit longer. But make no mistake, the commercial momentum is absolutely accelerating.

Okay. Excellent. And I was wondering, we've seen other technologies emerging in recent data center deals, small-scale gas turbines, gas engines. I'm wondering if you could describe what you're seeing with the competitive environment how your product compares to some of the other solutions? And just -- is the competition heating up? Or how do you see it playing out?

Look, I think the supply-demand mismatch is so large that everybody who has a solution that's viable today has a market out there for them to address. So you're going to see data center developers, hyperscalers wanting any and every solution that they can find. But there are very clearly, you're asking for distinction between us and other technologies. These were purpose built for the data centers. The additional benefit and value we bring to them is enormous compared to band-aided solution of something that was created for the mechanical age, trying to solve this very sophisticated digital AI problem. We stand to benefit every single time that we stand to benefit our end customer using our technology. And so you asked me to compare, let me compare other technologies that generate on-site create air pollution, we don't. Other technologies that now using mechanical combustion moving parts, cannot load follow and require lots of batteries to be able to maintain an on-site power because these are not connected to the grid, whereas our solid-state power does not require batteries and we are able to provide that power. Today, we are able to provide our power faster than most of the others who have supply chain constraints. We can expand our capacities a lot faster than anybody else. We are future-proofing our customers for future technology advances, whether it is in the field of DC power, whether it is in the field of carbon capture and zero carbon or a green molecule, we offer all those optionalities that others don't have. You put all those together from -- and then if you take the same amount of gas that is available, we can produce a lot more power and allow the hyperscaler to put out a lot more tokens. At the end of the day, it is converting those watts to tokens is where the game is, did the same amount of gas that's available, same amount of space that's available, we can produce a lot more tokens for the hyperscaler than any other technology can today, end-to-end. And so the value for the hyperscaler is not about the cost of power. It's about that cost of the entire value chain across the board. So price performance ratio we can compete with anybody. So that's the answer, David. Thank you.

Your next question comes from the line of Chris Dendrinos from RBC Capital Markets.

Congratulations on the strong quarter. I wanted to follow up on the Brookfield partnership here. And I'm hoping you could just expand a little bit on the relationship and provide some more details around the potential development time line? And then just how should we think about this partnership financially and how that benefits you?

Chris, thank you so much. Brookfield, look, they're an incredible partner to Bloom, right? I mean, they are at the heart of the AI value chain. And I think I mentioned this in the script, they've already invested over $50 billion in AI and want to triple that very quickly in the next two to three years. But put that in a broader context. They're one of the world's largest infrastructure owners with over $1 trillion in assets that comprise of 140 data centers operating and using approximately 1 gigawatt of critical load capacity and wanting to accelerate and grow that enormously in AI. On top of that, they have a portfolio of factories. They have a portfolio of commercial offices and real estate. There are all going to be beneficiaries of AI. And as they automate as they bring robots and those factories are going to need more power. So Brookfield is using their balance sheet and using their relationship with us as the power provider and making us the preferred choice that they would recommend to all their portfolio companies, including their data centers.

On top of that, Brookfield believes that they themselves are going to be a large AI infrastructure developer. And there they're going to use us. On top of that, if there are Bloom-sourced deals that require financing so we can offer a customer a PPA, they are willing to step in and be the financier for that. This is all not -- and they have made it very clear that this $5 billion investment is an inaugural investment.

Now what Brookfield -- with Brookfield, we have already done some deals together. And they have said that they will announce a European AI insurance data center before the end of the year using Bloom as a power source, so stay tuned for that. So it's a very big relationship. I cannot understate how important it is to us.

Got it. And I guess maybe just as a follow-up, sticking with the European opportunity here, can you maybe just expand on the global opportunity? And are you seeing the same kind of power limitations globally as you are in the U.S.? And does that present a strong opportunity for more international growth?

Yes. Chris, that's a great question. Look, I have been to these capitals, whether it is Frankfurt or it's Munich, whether it's Dublin, whether it's Taipei, okay? They all have a power shortage problem. And they all clearly recognize that their central power plants along with transmission distribution cannot keep up with AI speed. That's across the board. This is true in Delhi, this is true in Mumbai, okay?

So now what is happening in Europe, Asia, if you take as an example, I was just recently in Tokyo. And what I heard there is finally, the sentiment of natural gas, not just being a short-term bridge, but a long-term solution. And the agreements the U.S. is reaching with our friendly countries, our friendly allies to say we will supply you long-term LNG is now making them take a very different look at natural gas. And once that policy unlock happens of saying natural gas projects can move forward. we think there will be a tremendous acceleration in those places, and we are extremely well positioned to be able to play and the interest in Europe for our carbon capture solutions where you can go to almost net zero using natural gas as a fuel, tremendous interest there. No other technology, the Turbans and the Indians cannot do that. We can. So tremendous interest there.

Your next question comes from the line of Manav Gupta from UBS.

Generally, when you tell a customer, I can deliver the order in 90 days, the customer is happy to get the order in 360 days. So incredible feed delivering it in 55 days. My first question, sir, here is last week, Energy Secretary Center draft a proposal to FERC that would limit the regulatory review period for data center connections to power grid to just 60 days, expediting a process that can currently extend up to years. Help us understand how this could help Bloom Energy.

Manav, thank you for those kind remarks, a shout out to our team for doing that. So look, the first thing I want to say about that or people who are not familiar with that is, again, the Energy Secretary asked FERC to start a hearing process and a rule-making change to allow large loads like data centers and AI factories and other factories to get rapid interconnection with the grid, which has been an issue. And first and foremost, we applaud the policymakers and regulators for doing that. I think it's the right thing for us as a country to do.

The second thing I want to say is, if you read that announcement, it's very obvious even when you get that interconnection, they state very clearly, you're entering the age of BYOP, bring your own power, okay? You get curtailed even if you have an interconnection, if you don't bring your own power. And large AI data centers are not going to operate in a place where the utility is going to curtail them and not curtail them depending on what their load and peaks are, right? So that becomes extremely important. So obviously, time to power is the reason this is being done, we are able to provide our servers very quickly to a utility who wants to interconnect and offer the power to either a data center or a factory. And it's not unlike what AEP is trying to do. We think it's just going to accelerate other utilities wanting to do the same thing. So that's how we think it's going to help our business.

And for people from the utility listening to that, right, it's very simple. You don't make your nuclear power plants. You don't make your gas turbines, you don't make your fuel cell. You can now buy our fuel cells very quickly and install it in front of the meter and offer it to your customer. But when you do that, here are the additional benefits you get:

Ancillary support for the grid. Let me explain this. It's not an easy concept to understand, engines and turbines can offer reactive power, it is something that you can provide as a byproduct in addition to supplying power to your load almost for free into the grid, and the grid benefits from that in stabilizing the local grid.

Now engines and turbines can -- or any other kind of combustion device that as rotation or movement can only do that in a very narrow range. Bloom has an amazing range in terms of that power factor and that benefit will be enormous in places like [ PJM ], where you have grid condition and grid instability or places like California, where the amount of renewables you have completely destabilize the grid. We are a stabilizing factor.

In addition, we can easily provide a lot more power into the grid in short notice to make up peaks and nonpeaks if they want a net meter either as a utility because we are constantly standing and hot standby and when the data center is not using it, you can export that power. However, if you're going to use a bunch of turbines to be able to do it, you don't keep the turbines on hot standby, it' not economically possible. So you have to start it up and bring it up, which means you need 5, 6 minutes, and that may be the time you need to peak.

So we are a tremendous asset to benefit, first, the data centers being built fast, again, for the utilities to be able to provide that and win with that. Third, for the utilities to use our ancillary services and benefit from that. And fourth, because we don't pollute the air via a benefit to the communities where they are installed. So this is a win and when we love this proposition.

My second question, and I apologize in advance. I am an electronics engineer, but it's been two decades since I graduated. So in case it's an invalid question, please just ignore it. Sir, I recently read somewhere that some chip makers are looking to move from 400-volt AC to 800-volt DC by 2027. I think it was NVIDIA, I'm just trying to understand if that does happen, would it make your fuel cell even more efficient would DC/DC power be even more efficient than a DC/AC power because of transmission losses. If you could just talk about that.

Don't undersell your technical knowledge, it is spot on. So I think given how you phrased it, let me -- this is such an important question, and we didn't address it. It is probably a miss on my part to have not addressed it in the script. So let me take a few minutes to explain this so everybody understand how it's so important.

Here's the important part, and I want you to understand this. This moving from converting that 400-volt AC to 48-volt DC, which is how server racks, which are the size of a refrigerator roughly sitting in a data center. These are the machines that manufacture the intelligence, okay? They -- today, the standard has been ever since we had data centers, low voltage, like 48-volt AC. So it gets converted.

So think of this, the power is going -- so I'm thinking of an analog as you asked me this question. It may not be perfect, but I think it will answer the question. So think of power coming in into the rack as a hungry human being drinking water, okay? They can only drink through a straw. And that straw was sufficient. That is the 48 old DC because it's small wire to which a small amount of water comes into the store, that water was sufficient to satiate the thirst. That was when CPU racks were 13 kilowatts. We have put a lot of band aids on it to make sure [ Blackwell ] chips that come somewhere near the 130 kilowatts can handle it to the trough. Guess what? The [ Rubin ] chips and going forward are going to be 5x to 10x that. There is no way you can pump that much of fluid for the body to keep up if that's the amount of water you need, that's the amount of power you need through that little strong. But in the rack, there is no more space than the straw. What does that mean? You have to increase the pressure of the water that you're shoving to that straw. That's the only way you're going to get more water through and that pressure equivalent in the water is voltage in power.

So it is -- the loss of physics dictate that you have to go to an 800-volt DC architecture, if you want AI chips that have more power density, which is the only way you can improve upon AI in the next generation. This is not an if, this is not a nice to have. This is a must-have.

Now go to the other side. All our wonderful legacy power generation systems that helped us propel into the mechanical age was built for the mechanical age. They are like Niagara Falls dropping water and you cannot make it to the voltage that you need. You need to make it very, very high voltage. Otherwise, you can't bring all that water in a pipe of reasonable size to where you want to bring it. Even an on-site 50-megawatt turbine, cannot produce directly at 800 volts or the amount of copper you need becomes too bulky, too big. It's not just physically viable.

Guess what we did at Bloom? We saw this coming one day, we didn't know what day in 2000 when we initially created architecture. We built an architecture where we can feed these draws appropriately right at that 800 volts, and we decided every unit we have shipped for the last 15 years has that. But after that, we have one other box that takes that DC and makes it into AC and provides it because we were making color TV images, the world was only consuming black and white. So we are converting our color TV images to black and white but the other guys create black and white, and now you have to colorize them all and provide low definition when we already have high-definition color image. That's the analog.

So we are super excited about this. As I see it, it self evident to me that this has to become the standard and Bloom has set the standard for the digital age, digital power.

Your next question comes from the line of Nick Amicucci from Evercore ISI.

I just wanted to build upon on kind of the doubling of capacity by the end of 2026 and kind of the commentary that we would support ForEx, the fiscal '25 revenue. How should we think about kind of the utilization on that capacity as we kind of enter into -- again, as we enter into 2027, and we have that the 2 gigawatts kind of up and running. I mean because if we're exploring opportunities to go beyond that 2 gigawatts it seems like ForEx full year '25 revenue, that seems like a big number that we could get there relatively quickly. So I just wanted to parse that out a little bit.

Yes. So here is a simple way to think about it, right? We didn't get to where we are today to deliver what I just explained, this purpose-built factory based on just meeting a market demand as we see it right now, we just prepared ourselves. What is the beauty of Bloom being able to expand its capacity and offer what we do is the return on investment like invested capital. So we are fiscally very disciplined, and we only make decisions based on that added cost and its absorption will it have a great rate of return.

So we have a very disciplined process on this. And on top of that, we have a very clear understanding right now given time to power shortages and the importance of this as a nation state issue for AI. We are committing to strive and work as hard as we need to and stay ahead such that we will never be the constraint to our customer on growing their data center. That's what we are positioned for and we will increase capacity, we will increase it in whatever steps necessary as we see fit but as you saw, this 2-gigawatt capacities, all systems go based on that. Would we use it for the capacity, when we use it, will we use it for steady capacity all that, you'll hear from us as we talk about our backlog and other things next year. But we are now using our OpEx wisely to invest in capability and talent to think about how do we expand beyond 2 gigawatts. That's all I can say right now.

Got it. That makes sense. And so as we see kind of the here and now, obviously, the power demand is here and obviously, you guys are ready, willing and able to address it. But as we kind of think out and I bring this up because you had mentioned an inference data center in Europe, I guess just -- can you just kind of convey the additive value when we think of kind of the inference and when latency becomes an issue when we get to inference and reasoning within the AI complex, how kind of Bloom are taking that?

So thanks for asking that question. It's very important, right? Here is the beauty, our exact same architecture with fewer LEGO blocks, okay? Think of each of our Bloom Power Systems as a LEGO block with fewer LEGO blocks is an inference data center, multiply that many times over, it can power a training data center. No difference, know the technology can do that. That's how we built it. That's the power of our modular fault tolerant architecture, number one.

Number two, inference data centers are going to be close to your bedroom window and your office window. You don't want that to be polluting, you don't want that to be noisy. We are -- we should be for those inference data centers, the power producer of choice.

Your next question comes from the line of Ben Kallo from Baird.

K.R. Maybe could you just talk about, I think, the biggest project you guys have announced is 80 megawatts with SK, could you just talk about how you view -- because there's been permits out there, huge numbers on them on doing bigger projects, how your customers gotten comfortable with your technology over time? And maybe the time to power how you think about the size of projects you can do and where we think you guys fit in, if it's 100 megawatts or 900 megawatts.

Ben, that's a very good question. And look, Again, our architecture was purpose-built. Our factories to copy exact modules and the boxes don't know whether they are sitting along with 5 other boxes or 5,000 other boxes. Nothing in our scaling at scaling risk, right? So yes, we are talking to customers with a lot bigger stamps right now and working with them on ideas and projects in various stages of negotiations of much larger sizes. We can do that, and we can do your neighborhood retail store, and we are talking to customers about that, too. So that is the flexibility of our architecture. You're adding no additional risk. In fact, think about it because these are part swappable LEGO blocks, the more LEGO blocks you have, the more reliable our system gets. So large block power becomes a lot more reliable than small block power. Thank you.

Maybe a follow-up. Just as we think about capacity, I think other people asked about this, but we see these big numbers out there. what's go-no-go decision from going 2 gigawatts more? And how fast can you do that?

Look, we get a stuff. You're building capacity too fast. We get accused of we don't think you can build too fast. Let me be very clear. We are going to strive to make sure we are able to provide power for our customers before they are ready for it. We will not be the bottleneck. And as we designed our factories, we built it with that in mind. Thank you.

[Operator Instructions] Your next question comes from the line of Mark Strouse from JPMorgan.

So K.R., things obviously are changing very rapidly here. it's been a bit [ there ] since you've provided kind of long-term margin targets. So kind of to the earlier point about capacity expanding but also as the utilization of that capacity increases how we should think about kind of gross or operating margins under that scenario?

Thank you. So look, here's how I'd answer it. [ Retail 90 ] days from now to hear our annual guidance for next year. But in the meantime, if you want to think about it, here's how you can think about it. For over a decade, every single year, we have a cost down in double digits, number one.

Number two, when we transact on electricity as we grow our volume, the pricing pressure on electricity is going to be based on the macros and there's a shortage of electricity. So you go figure out what that pricing pressure would be. And we have tremendous operating discipline within the company that you have seen us exercise these last 3, 4 quarters, and that's how we'll continue going forward.

Our factories are not capital intensive. We have made that statement very clear. You know what those numbers are. There is it that we will spend, investment. We'll invest in our people. We'll continue to invest in our technology. We will invest in the talent necessary for scaling our operations. We'll invest in the skills needed in our commercial team to go capture opportunity. And we will continue to invest in technology to further enhance our leadership position on on-site power. So those are the things to take away.

Your next question comes from the line of Michael Blum from Wells Fargo.

You've given us some good information on how to think about the Brookfield partnership and the scope of that. But can you give something similar with Oracle? Can you give us a sense of the size of that opportunity set? And how exactly Bloom will play a role in that partnership because they also obviously have pretty big ambitions as well?

Michael, I can't speak to any one customer. You should be asking them about that question. But I think what I can refer you to there would be their statement when we had the press release saying that what we did for them earlier was the first of many, right? So we think they are going to play an extremely big role in this space, and they are going to be growing in many, many geographies. And they are not only looking at how we have executed on the first year, but they're intimately familiar with what's in our technology road map and all the value we can bring to them. So we just obsess on placing the customer, then the customer will do the right thing. So that's what we do. Thank you.

Your next question comes from the line of Ameet Thakkar from BMO Capital Markets.

I've just got a quick kind of housekeeping question. Your 10-Q kind of refers to $288 million of related power sort of related party revenues during the quarter. I was just wondering, is that related to Brookfield or is that related to SK?

Yes. This is Maciej. Part of the contracts with Brookfield, we made equity investments into the vehicles. And because of those equity investments, those JVs became a related party to grow. And that's what created the disclosure around the related party revenue.

And again, it's important to note that the equity investments are fairly small. The way the contract is being set up is that we put a little bit of ceiling on those investments to be very significant. That said, there's a criteria to go through. And if you meet the criteria of equity investment, you get into the related party disclosure.

Your next question comes from the line of Colin Rusch from Piper Sandler.

No, it's Colin Rusch, but I am from Oppenheimer. So just in terms of the balance of 2025 and looking into '26, can you talk a little bit about the mix shift from direct product sales into some of these financing options with partners or related parties. And then also if we could get a quick update on the CFO search and that coming to conclusion?

Yes. I'll take the CFO search. Look, it's a very important position for us. We take it very seriously. And so we have a process in place and that search is going on. We have a sense of urgency, but no sense of rush. So we will let you know when we hire one.

Yes. As far as the financing goes, if you go back and we talk about this in pretty good detail in our 10-K, and where -- there are three ways of going to market. There is a direct sale, which we refer to as CapEx, which is effectively a customer showing up and writes a big check, which effectively [indiscernible] prepaying electricity or a number of years as a PPA financing structure in place and the managed services, which is the sale-leaseback transaction. We haven't done many service transactions in quite some time. We don't expect to do those going forward. And I would say majority of the transactions we get into are actually further PPA structures, although there are some topic deals every quarter from time to time or the customer is wanting to find [indiscernible].

Your next question comes from the line of Maheep Mandloi from Mizuho.

Thanks for taking the question, most of the high-level are answered, but [indiscernible] housekeeping on the guidance for Q4, if you could you just talk about like the reason to not disclose that? Is it just timing of these lumpy installations you have in December or January or something else over there?

Yes. I think you just answered your question, Maheep. We have said that at the beginning of the year, why do we give a range? Because project-based installations, many of them being greenfield for changes and changes in variations. Lots of things could happen on the customer end. We have no difficulty supplying our boxes on time, ahead of time, as you saw but the customer has to be ready to take the power, which is when we ship. So a project can fall a few days in front of or on the other side of December 31, which is just a deadline, and it has no impact on that project or that revenue other than a pure timing issue. And yes, with 365 days in the year, we would give you on guidance. Now 300 of those are gone, and we have 65 days left. So we're giving you a slightly better guidance, but we can't pinpoint it. That's exactly right.

Your next question comes from the line of Sherif Elmaghrabi from BTIG.

For the Brookfield partnership, you mentioned they're bringing a substantial balance sheet to the equation. But any capital commitments for Bloom under those joint ventures or other costs related to that, obviously, besides [ Fremont ] CapEx that you're already investing?

Yes. Other than the equity investments that we agreed to make on the core respective projects, very small equity investment, there is not.

Okay. With that, I just want to bring this to a close as we are getting to the top of the hour. Thank you all. We are delighted with our results in Q3. Our commercial activity is robust and the momentum is accelerating across the board. This year has been a big transition year for Bloom. I want to thank our team inside for stepping up and contributing to our success in such a great way. We appreciate the trust of our long-term shareholders, and we welcome our newest shareholders, and thanks to all of you for placing trust in us. And I think we have an exciting journey ahead of us together as we go forward to become the standard for on-site power. Thank you, and have a great day.

This concludes today's conference call. You may now disconnect.