Operator
Welcome to the call. Good morning, and welcome to the Amprius Technologies Third Quarter 2026 Earnings Conference Call. Joining us for today's presentation are the company's CEO, Tom Stepien, and CFO, Ricardo Rodriguez. At this time, all participants are in a listen-only mode. Following management's remarks, we will open the call for questions. Please note that this presentation contains forward-looking statements, including but not limited to statements regarding the company's financial and business performance, business strategy, future product development or commercialization, new customer adoption and new applications, the company's growth and the growth of markets in which the company operates, and the timing and ability of Amprius to expand its manufacturing capacity, scale its business, and achieve a sustainability These statements involve known and unknown risks, uncertainties, and other important factors that may cause Amprius' results, performance, or achievements to be materially different from any future results, performance, or achievements expressed or implied in such forward-looking statements. For a more complete discussion of these risks and uncertainties, please refer to Amprius' filings with the Securities and Exchange Commission. This presentation includes non-GAAP financial measures, including adjusted EBITDA and adjusted EBITDA margin. Our non-GAAP financial measures do not replace the presentation of Amprix's GAAP financial results and should only be used as a supplement to, not a substitute for, Amprix's financial results presented in accordance with GAAP and may not be comparable to calculations of similarly titled measures by other companies. A reconciliation of adjusted EBITDA to GAAP net loss, the most directly comparable GAAP financial measure, is included in our press release, a copy of which is furnished to the SEC and posted on our website. Finally, I'd like to remind everyone that this conference call is being webcast. A recording will be made available for replay on the company's Investor Relations website at ir.amprias.com. In addition to the webcast, the company has also posted a press release that accompanies these results, which can also be found on the Amprias Investor Relations website. Before turning the call over to management, I want to highlight a few near-term IR events. Amprius will be attending the Craig Hallam Conference on November 17th in New York and the B. Reilly Conference in New York on December 2nd. We hope to connect with many of you at these upcoming events. I'll now turn the call over to Amprius Technology CEO Tom Stepien for his comments. Sir, please go ahead.
Welcome, everyone, and thank you for joining us this morning. Ampryus continues to see strong demand for our energy-dense silicon-anode lithium-ion batteries from a growing set of aviation, defense, and advanced mobility customers. In the third quarter, that robust demand resulted in record revenue of $39.1 million, up 83% year-over-year and 15% sequentially. As shown on slide 3, our Q3 revenue represents an annualized run rate of over $156 million. Given our double-digit revenue percentage increase in Q3, A rapidly growing order book, along with an expanding group of contract manufacturers that support our capital-like business model, we are raising our full-year 2026 revenue target to $145 million, up from at least $140 million. This is the third consecutive quarter that we have increased our outlook. We saw particular strength in North America this quarter, with revenue in the region up 2.3x compared to last year. North America represented 31% of our business in Q3, up from 19% in Q2. Europe remains our largest market at nearly 60% of overall revenue. The third quarter was an eventful period for Ampheus, which included product introductions, new manufacturing relationships, and a growing customer base. Let me touch on a few highlights. In September, we introduced Sycor 500, a new 500 watt-hour per kilogram cell designed for long endurance aviation applications. Previously, producing these extremely high-energy-density cells required custom manufacturing tools, specialized material handling equipment and procedures, and non-standard cell designs. SciCore 500 is markedly different, designed for commercial lithium-ion processing equipment and standard cell formats. Making it compatible with Ampereus' global contract manufacturing network. We believe these new cells are ideal for the sustained low-rate discharge profiles required for high-altitude platform systems, or HAPS, as well as for fixed-wing drones and other long-endurance applications. In HAPS, Stored energy and battery weight decide whether persistent stratospheric flight is possible. Solar-powered platforms charge with batteries during the day so that they can stay aloft through the night. Higher specific energy cells increase flight time, range, and payload without adding weight. We plan to ship SICOR 500 cells to HAF's companies in the fourth quarter. They support extended-duration missions where platforms stay airborne in the stratosphere, between 60 and 65,000 feet, for weeks and months at a time. These systems provide broadband communications, bringing Internet access to remote areas. HAPS also enables Earth observation and environmental monitoring for tracking methane or other greenhouse gas emissions, for example. I want to highlight one of our European customers, Quantum Systems, a leading German developer of autonomous unmanned systems. Quantum is growing rapidly, attracting world-class investors, and expanding its portfolio across air, land, and sea. Ampryus has supplied silicon anode cells for Quantum's Vector and Twister ISR platform for several years, and we are proud that our technology helps deliver the endurance these missions demand. Relationships like this one reflect the durability of our customer base and the growing role of high-energy density cells in allied defense programs. In Q3, Amprix entered new market segments, defense and maritime vehicles. We began shipping SciCore cells to a leading European manufacturer of autonomous deep-sea underwater vehicles that operate at depths of more than 3,000 meters for commercial, scientific, and defense applications. These vehicles execute lengthy missions and are far away from any opportunity to recharge, so every watt hour on board really matters. That makes them a natural fit for our high-energy density silicon anode cells. This is the first time SciCore technology has gone beneath the sea, and we see it as a promising new avenue for growth. Let's now turn to the supply side. In September, Amprius was selected for a competitive contract from the U.S. Department of War's Industrial Base Analysis and Sustainment Program to retrofit an existing domestic lithium-ion battery production line. The retrofit changes one portion of an existing factory, converting production from standard electric vehicle batteries to our advanced battery systems that power Group 1, 2, and 3 drones. The contract totals $75 million, $22.3 million of which has been obligated. Our partner in this contract is a major South Korean battery manufacturer that operates the existing U.S. line today. At completion, expected in early 2028, we expect that the line will have the capacity to produce 12 million Ampereus pouch cells annually. Importantly, these cells are aligned with the domestic sourcing requirements of the National Defense Authorization Act. The conversion project is expected to cost approximately $100 million, including in-kind investments from Ampryus. This project will significantly increase our ability to provide standardized, NDAA-compliant, high-performance pout cells for key customers at scale. As you are aware, Ampryus has chosen a capital-efficient production strategy. We have long worked with contract manufacturers in China and South Korea, and we have been expanding our work with partners in the United States. For instance, we recently entered into a contract manufacturing agreement with Kamenus, a U.S.-based manufacturer of lithium-ion battery cells and electrodes, to build our pouch cells at its facility in Indianapolis. The agreement spans the cell production process from partial assembly of Ampris-supplied components to full cell production and advances our development of a resilient NDAA-compliant supply chain. I want to reiterate that Ampris is shipping NDA-compliant cells today, and we remain on track to reach full NDA compliance with domestically produced cells in 2027. Please now turn to slide four. As the unmanned aerial vehicle market matures, we see segmentation of the vehicles and applications. We have mapped various drone applications with Ampria cells that enable the performance these applications demand. Our catalog covers UAV missions from minutes to months for both one-way munitions and reusable platforms. Reusable drones are shown at the top of our diagram. Intelligence, Surveillance, and Reconnaissance, ISR, drones, drones as first responder, DFR, and delivery drones tend to need a combination of power for takeoff and landing and energy for flight time. AMBRIES meets these requirements with a balanced line of side-core cells. Haps. Like those described earlier, we require maximum energy, which we deliver with our new Sycor 500 cells and other batteries depicted in red. The bottom of our diagram covers low-cost, attritable, or one-way drones. Our balanced cells are often used by deep-strike drones. Interceptor drones need power to quickly reach maximum speed and intercept a hostile drone and our high-power cells deliver. Cell content per aircraft ranges from hundreds of dollars to tens of thousands, so mix matters here as much as volume. Please note that Amprix's silicon anode cells deliver premium performance. That is not required by all drones. The power needs of low-cost first-person view, FPV drones, are often met by off-the-shelf, low-cost graphite anode cells. We do not chase these opportunities. Please now turn to slide 5. While the results of the recent U.S. midterm elections could have an impact on the fiscal 2027 U.S. defense budget, We believe that the widespread adoption of military drones by armed forces across the world is here to stay. As we have noted before, the Trump administration's proposed 2027 budget calls for more than $50 billion in outlays for the Defense Autonomous Warfare Group, DOG, an arm of the Department of War focused on drones and related hardware. While the U.S. Congress has yet to pass the 2027 spending authorization and the actual budget might be smaller than the original request, it is nonetheless clear that the Pentagon will be making a major and growing outlay for autonomous capabilities for years to come, and Amprix expects to be a beneficiary of that trend. Independent of the size of the 2027 defense budget, we believe the U.S. will meaningfully increase spending on drones. Emil Michael, the Department of War's Undersecretary for Research and Engineering, said at the Wall Street Journal tech conference last week that the U.S. was, quote, caught flat-footed by drone warfare and weren't prepared because we just didn't imagine it. I just want to make sure we are imagining the next fight, unquote. Many countries recognize the asymmetric advantage drones provide on the modern battlefield. U.S. allies, including Ukraine, Taiwan, and the U.K., have rolled out drone programs with combined spending of more than $20 billion over the next three years. Other programs are underway in Germany. Poland, the Nordics, Japan, and South Korea. Note that several countries either prefer or require allied source sales, which favors suppliers like Ampryus. We are keenly aware of these trends, and we are positioning Ampryus to win. In Q3, we hired two Europe-based sales account managers. On Monday, November 9th, we hosted a high-energy battery day in Tel Aviv, attended by about 100 people from dozens of Israeli companies. On slide 6, we wanted to highlight our view of market share and growth rates. AMBRIUS estimates the total defense-related sell demand in 2026 from the U.S. and allied nations is $445 million, up 28% from 2025. With our updated 2026 revenue forecast of $145 million, nearly twice our 2025 total, we believe we are growing approximately three and a half times faster than the market. Our industry-leading technical performance has allowed us to outperform the overall market. Looking forward, by 2030, we expect loitering munitions and interceptors to be the fastest-growing segments with a projected 58% compound annual growth rate. These applications align well with the core strengths of our cell portfolio. Let me now turn the call over to Ricardo to review our Q3 results in detail.
Thank you, Tom. Good morning, everyone. I'll start my remarks with slide 7 as the backdrop. Before we dive into the results, I'll summarize the main things to take away this morning. First, we had a record quarter. Revenue of $39.1 million, up 83% year-over-year, with gross margin holding at about 27% through a tougher mix with more cylindrical cells than originally expected. Second, gross margin has the potential to improve from here. I'll walk through each driver of margin this quarter and why I see each one as an opportunity rather than a headwind. Next, behind our third revenue guidance race, we've made deliberate investments to keep outpacing the markets that we serve. We are raising 2026 revenue guidance to $145 million. We are also updating adjusted EBITDA guidance to at least $2.5 million as we invest ahead of projected demand. And fourth, our balance sheet is actively funding the ramp. We have $60.7 million of cash and no debt. The cash we used this quarter went into securing supply for Q4 and early 2027, and we believe it will quickly be turned into more cash as we fulfill near-term demand. The third quarter was our seventh consecutive quarter of sequential double-digit topline growth. North America drove the quarter, more than doubling to $12.2 million, while Europe held steady at $23.3 million. Through nine months, revenue is $101.7 million, up 113%. Gross profit was $10.5 million, up 13% sequentially, and more than 3x year-over-year. Gross margin was 26.9%, close to the level with Q2's 27.3%. And up from 15% a year ago. For the first nine months of the year, gross margin was 25.1% versus 5% last year. Excluding Colorado costs in both years, gross margin was 26.3% year-to-date and 15% for the same period last year. A year ago, I said there would be a quarter or two where margins take a step back as we manage mix. And as North America grows ahead of us, getting paid for NDAA compliance. Both factors apply in Q3, but we still held gross margin at 27% by managing a couple of ins and outs. Two headwinds cost us about eight points of gross margin. Mix took about four, more cylindrical cells, much of it shipping against the large European order, and more shipments to North America. Freight took the other half of these eight points as we shipped by air quickly to meet demand. We nearly made all of it back, though. About four points came from lower expenses in Fremont, and an $800,000 portion of that moved to R&D, since that team focused on development and the engineering prototyping of the IU line. And the rest we were able to recover in pricing. Each of these drivers is a lever that our team is getting better at managing proactively. On the next slide, I will outline in more detail how these factors drive our margin outlook. The Q3 operating expenses were $15.2 million, up $1.7 million sequentially. Excluding the $800,000 of Fremont expenses that were reallocated from cost of goods sold, operating expenses grew by about $900,000. That is in line with the $1 million quarterly rate I outlined in August, and it went mostly to development, including Cycor 500. Our net loss was $3.5 million, or negative $0.02 per share, 10% narrower than a year ago. Adjusted EBITDA was negative $1 million, and negative 3% margin. Over the last 12 months, adjusted EBITDA was negative $1.4 million on $127 million dollars of revenue, and very close to breakeven. Now let me turn to our outlook on slide 8. In March, we set an initial 2026 revenue outlook of more than $125 million. We raised our forecast to $130 million in May and then to at least $140 million in August. With what we know today, we are raising it a third time to $145 million. That is $20 million above where we started the year. The chart at the top of the slide shows why. Our annualized run rate went from $114 million in Q1 to $136 million in Q2 and $156 million in Q3. Our guide implies Q4 revenue of about $43.3 million, up about 11% from Q3. Our goal remains the same as it has been all year, to deliver what we promise and when we can, some more. Our implied Q4 annual run rate would be of $173 million. On gross margin, we continue to expect at least 28% for the full year. We're at 25% through nine months, so the fourth quarter carries a real step-up. We believe that we can get there by managing the following drivers. The first is mix and volume. The orders that we have in hand for Q4 will likely carry a richer mix than Q3, with more pouch-sell demand in the current and expected purchase orders. The second is unit cost. We continue spreading fixed costs over more cells, and so this works to our advantage. The third one is freight. Freight was a painful 11% of revenue in Q3. Moving most cell shipments from air to ocean, starting around year-end, can half that cost, or provide us about 5 points at today's revenue run rate. That is mostly a 2027 benefit, but with more inventory on hand, we are going after this for the first time before the end of this year. The last one is pricing. In this high-demand environment, we'll continue optimizing for this, as the markets that we sell into confirm that we have a uniquely differentiated, high-performance product. Mix will keep moving from quarter to quarter as we adjust the shifting supply and demand trends. But each of these levers is in our hands, and nothing about our path to gross margins above 30% has changed. As I've said before, it won't be a straight line up and to the right, but our visibility is improving over time. Moving to the rest of our outlook, we now expect adjusted EBITDA of at least $2.5 million, and I'll walk you through that bridge on the next slide. We're reiterating a net loss of $10 million or less. That includes the $1.9 million non-cash warrant charge from Q2. Excluding it, the net loss is expected to be about $8 million. We are updating our EPS outlook to a loss of $0.07 or less per diluted share and $0.06 excluding the warrant exchange. These all assume a share count of 143.4 million shares. Finally, we continue to expect capital expenditures below $10 million for the year, funding DIU-backed investments in Fremont while keeping our capital-efficient, low-fixed-cost model. On slide 9, I'll walk you through our target adjusted EBITDA for the year. As I mentioned earlier, we are updating our 2026 outlook to at least $2.5 million from more than $4 million in August. The bridge is simple. The $5 million revenue raise at a gross margin of at least 28% adds about $1.4 million of gross profit. We have chosen to invest about $2.9 million more in operating expenses this year. These investments are already paying dividends in various aspects, such as R&D. Taking SciCore 500 at 500 watt-hours per kilogram on conventional lithium-ion lines from launch to customer qualification, and we also continue certifying our entire cell portfolio to more stringent standards that will keep extending our competitive lead. The second is enabling U.S. production. Good examples of this are starting production on the DIU pilot line in Fremont in December, and enabling Kamenas in Indianapolis for pouch volume production in the first quarter of 2027. And finally, driving our recent go-to-market wins, including the design wins that Tom described earlier in his remarks, including the sub-sea UUV win and the expected quantum systems work into orders. One last item to keep in mind for Q4. We expect to complete about $6 million of work under our $18.1 million Defense Innovation Unit Agreement before the end of the year. We record that as other income, not revenue. Consistent with how we have always defined adjusted EBITDA, it is included in our outlook. We ended the third quarter with $60.7 million of cash, no debt, and no at the market equity offering facility that would increase the share count. Cash was down $13.8 million in the quarter. Our operations used $7.8 million, and capital expenditures were $5.2 million, all at our Fremont facility and supporting the IU-backed investments. The operating cash use was a working capital built to drive near-term growth, and it was deliberate. Prepaid expenses rode $14 million to $18.9 million and included advances to our partners to secure supply for the Q4 and 2027 ramp. Receivables grew $5.2 million to $45.9 million with higher shipments. And inventory grew $4.4 million to $15.9 million, also in order to drive near-term growth. Higher payables and the crude liabilities offset a good part of that. This build is also what is now letting us ship by ocean instead of by air, which drives our long-term gross margin floor. Working capital ended at $107.8 million, compared to $90.7 million a year ago, and total stockholders' equity was $124.7 million. Our contract manufacturing model keeps capacity growth capital light, and our balance sheet funds the ramp ahead. As we get to the close of the year, I'd like to thank our team for delivering a record quarter, margin drivers that we can pursue and act on, A third revenue forecast to raise, and for protecting a balance sheet that funds the ramp with no debt. We expect to head into 2027 with more customers, very clear demand visibility, NDAA-compliant U.S. capacity coming online, and the cost structure built to capture more than our fair share of opportunity. With that, I'm happy to turn the call back to Tom for his closing remarks. Thank you very much for your attention and continued support.
Ricardo, thank you. Ampheus continues to see wide and varied growth opportunities across multiple markets and geographies. We remain bullish about the opportunity in drones for the defense sector, but that is far from our only market. We expect growing commercial demand for drones in delivery, inspection, agriculture, search and rescue, and other areas in the months and years ahead. We're equally encouraged by the outlook for our battery's e-mobility applications, including e-motorcycles and scooters, for personal and commercial robots, and for electric vehicle takeoff and landing aircraft, for point-to-point regional transport of both passengers and cargo. In short, the future looks bright for Ampheus, and we appreciate your continued support on this journey. Operator, we can open the floor for questions.
Operator
Thank you. If you'd like to ask a question, please press star 1 on your telephone keypad. A confirmation tone will indicate your line is in the question queue. You may press star 2 if you'd like to remove your question from the queue. For participants using speaker equipment, it may be necessary to pick up your handset before pressing the star keys. To allow for as many questions as possible, we ask that you each keep to one question and one follow-up. Thank you. Our first question comes from the line of Colin Rush with Oppenheimer & Company. Please proceed with your question.
Thanks much, guys. You know, you raised the target for 2030 in the deck from $600 million to $800 million in revenue. And you've talked a little in a lot of detail here around the drone opportunity. Can you talk a little bit about what's happening with both U.S. and allied military procurement and those sales processes that you're seeing that's going to support that increase in revenue over the next four to five years?
Yeah, one of the slides, Colin, thank you, this is Tom, shows the worldwide opportunity and the total $20 billion that you get over three years, and we expect that will come through and more will be added to it, and as we mentioned, some of that is a function of actual 2017 defense budget. But what we are seeing is flow through. The numbers that you see that we show on our slides are translating to orders that we are receiving here. The increase in North America is one indication of that, right, 2.2X, quarter over quarter. So it is flowing down to the drone manufacturers and to the battery suppliers like Ampryus that are supplying to the craft. It is happening.
Yeah, the other thing to add, Colin, is that you're referring to slide 17. That number has always been anchored by what we believe that we can realistically supply given our existing agreements and capacity plans, and so the $200 million add in 2030 capacity is basically what we expect to have up and running with our South Korean partner by then, and then also the capacity at Kaminas. So, between both facilities, we should be adding, depending on the mix, anywhere between 270 and 300 megawatt hours per year.
And then on the technology roadmap, you've laid out some of those targets and the progress that you can make over the next couple of years with performance. Can you talk a little bit about where you're at on executing against that roadmap, what that's doing for your customer relationships, and any update on various materials or chemistries that we should be thinking about that could start impacting some of those performance metrics?
We have a robust technology team. They are on the treadmill. Sitecore 500 is the latest example of an industry-leading cell that is now available. What we're trying to do is get earlier into the design cycle of our customers. So rather than wait and understand publicly what might be available, we're reaching out and sharing roadmaps under appropriate NDAs, of course, with customers like the one we mentioned in the call here with Quantum Systems. We're close with them, we're close with others, and that way we can share with them not just what was recently released, but what we're working on for 2027. You asked about the supply chain. We have a NDA compliance supply chain. There are 18 different components to our batteries, anode, cathode, electrolyte, separator, etc. We have reviewed those. We have individual components for NDA compliance with the Department of War. They are happy with it. We are in execution mode on that. We're supplying those cells today with some of the partners we mentioned. We have more to do, of course, but we're happy with the roadmap on the technology side.
And the timing wise, Colin, yeah, I mean, basically the Cycler 500 announcement came out three months before we promised it when we first put out that chart on the evolution of the performance of the cell. So we're on track.
Perfect, guys. Thank you so much.
Operator
Thank you. Our next question comes from the line of George Gianaricus with Canaccord Genuity. Please proceed with your question.
Hi, everyone. Good morning, and thank you for taking my questions. Hey, Ricardo. So maybe to focus a little bit on your long-term margin guidance for 2030, as you're having these conversations with your contract manufacturing partners, how have those sort of either increased your conviction or maybe decreased them in those margin targets over the next five years? Thank you.
Yeah, so, you know, one thing that gives us even more conviction about the longer-term gross margin target is that as our volumes increase, we have a better idea of what the fixed-cost structure of the CMs looks like, and so one of the points that we've been making to them here as the volumes have grown is Okay, given that those are fixed, why don't you just charge them to me as a fixed expense, and let's keep the variable piece of it variable, and that way we can share in some of that operating leverage. We're also developing capabilities in-house to manage the CMs, so some of the setup and engineering work that the CMs would normally be doing themselves and amortizing within each piece. This will ideally no longer be there with some of these new relationships that we're setting up. I mean, we're getting pretty hands-on at owning the process and driving this within the CMs, and as we do that ourselves, we think we can find scalability that otherwise wouldn't be there if it's kind of bucketed within each CM.
Thank you. You know, and as your market opportunity grows, there are competitors trying to break in as well. And what would you say the factors are that most frequently tip competitive evaluations in your favor?
Well, the first one is really just the ability to fulfill the order, right? You know, We have seen a lot of competitors put out press releases where they're, in essence, taking the specs of our previous generation product, thinking that that will be enough to get volume. But they really have a chicken or the egg problem, right, where they don't have the customer. The technology is, frankly, a generation behind what our team is currently developing. And, you know, we've seen several of our customers actually place relatively larger volume orders with these potential competitors only for them to not be fulfilled because the product only exists in a slide.
We also have the capability here, George and Fremont, with our pilot line. It'll be fully up and running at the beginning of the year. There's 30 new tools, 28 of which are installed. We still need a couple more to come in. That allows us to stay on that innovation edge, and we can make 700, 1,000 cells a day once we are at steady state. And we do that not for volume orders, but for new orders and different chemistry platforms and customers that want us to turn the knobs and provide new cell sample cells. So that ability to do quick turn designs directly with our customers rapidly helps us stay where we want to stay.
Yeah, I think if you look at, for example, slide six and the demand that Tom framed out, we're talking about, you know, roughly a $450 million sell market here in 2026. And, okay, we're, you know, selling our $145 million of revenue roughly into that, assuming that all of our sales are into this sell market, which... Only 75% of them really are. But when I look at the competitive landscape, you know, the second and third biggest competitors at most can do, you know, $80 million of revenue per year, and the second one can do, you know, $40 million a year of revenue if they execute. And also, they're capacity limited. So I do think that, you know, we have plenty of room here to grow while the competition is kind of capacity constrained, and that's why we're finding out that this contract manufacturing model, if we can stand this capacity up quickly enough, is actually pretty good.
Operator
Thank you. Our next question comes from the line of Mark Shooter with William Blair. Please proceed with your question.
Hey, Tom and Ricardo. Congrats again on the quarter. Touching on the new slides, of course, touching on the new slides that you had of breaking out the Allied defense drone spending. Related to the midterms, as you mentioned, Tom, I think the common through line is that maybe the blue wave will tamper defense spending. But again, I don't think we agree that that would impact the current U.S. drone demand. But to look at it at a slightly different angle, if Democrats want to increase the support for the Ukrainian defense efforts. Could the blue wave actually be more of a tailwind, and not even net neutral or net negative, but could the blue wave be a tailwind in 27?
Look, it certainly could be. We all want that conflict to end, of course. So there are that possibility. And it's also – The slide before the one you mentioned that showed the different types of drones, super interesting to us, because if you look at that, all right, maybe some of the one-way drones would decrease if peace does come sooner than, and of course, we all want that. But the surveillance and other ones will probably increase in a peace-type situation. Thank you. We believe that there are a lot of legs to this drone story. It's the decade of the drone, the tides coming in, all the euphemisms that you want, again, because of the asymmetric advantage. Almost independent, and we tried to make this point, of the actual budget, drone spending will be higher, right? Fewer tanks, more drones in defense, and then also we see some of that same asymmetry with delivery. We've seen recent announcements from companies with agriculture, etc. So, it is a very good spot for us as a battery supplier to drone. Companies across the different segments and across the different geographies.
Thanks, Tom. And on that next slide, where you break out the sell demand and the market size, I did notice that pricing was at $350 a kilowatt hour. You know, our understanding of pricing is almost double that. So I'm just wondering if those numbers that you have for the TAM could be conservative, or are you assuming that there is some pricing decline?
They are. This includes a lot of cylindrical cells as well, so I think it's worth keeping that in mind. The $600 plus, and in some cases even $1,300 per kilowatt hour, is at the pack level, so there's some add-on to this. Here, we wanted to just strip it down to the cells, and we were, you know, pretty conservative on what the mix of this would be and also on what the pricing could be, but that's a good catch, Mark.
Operator
Thank you. Our next question comes from the line of Eric Stein with Craig Hallam Capital Group. Please proceed with your question.
Hi, Tom. Hi, Ricardo. Good morning.
Hey, so I do want to go back to slide six. And, you know, you just talked about the limitations of your competitors. And obviously, you've got the competitive advantage and the broad product offering, etc. You know, when you right now share, you know, estimated share around 30%. I mean, how do you how do you think about share, you know, as the market? It continues to expand pretty rapidly. I mean, is this growth with the market? Because you can actually deliver on this, are there market share gains? How do you think about that, I guess, 2028, 2030, and beyond?
Yeah, that's a good question. For 2028, the recent capacity increase announcements definitely help us go and maintain our share. But I would say stay tuned, right? Obviously, we have to assess where the demand is coming from. All signs point to it continuing to come from Europe. And so, you know, we haven't announced anything around capacity in Europe. I think that the roughly 300 megawatt... Hours that we'll add here in the U.S. over the next, you know, year and a half, two years for NDAA compliance help us satisfy this near-term need that we see. And if you look at the U.S. portion of the bar as a floor, it, you know, it can more than use up that demand that we're setting up. And so, yeah, I think this will evolve. I mean, we'll look at these market figures quarterly. We're pretty in tune with the main demand drivers, and I think that's the beauty of our model, where we're, in essence, building up capacity much faster than what it would take to set up our own captive plant. I mean, even the... The one plant that would produce the number of SKUs that we have in a mix of pouches and cylindricals at these relatively low volumes, right, and I say relative to other markets like EVs and stationary storage, the plant just does not exist. I don't think anybody, and as we talk to our experts around the process design and how to get this done, I don't think anybody can set up one plant that is, you know. Sub a couple gigawatt hours and produce so many different types of cells with these margins. It does need to be done through multiple leveraged facilities at contract manufacturers. And the speed is also critical, too. I think the speed is what's enabled us to get more than our fair share of this market here. And if we keep going, it's what will enable us to keep outpacing the growth of the market as we laid it out here on slide six.
Got it. Thank you for that. And maybe just for my follow up at the end of the prepared remarks, you know, obviously bullish on drones, but then talked about commercial and talked about robotics and light EVs, etc., other applications. I mean, maybe I would assume 2027 is still. Vast majority is going to be about drones, UAVs, but maybe as you think about the mix evolving as you get out to 2030 and beyond, how do you think about that in terms of your overall business?
Yeah, 2027 will still be drone dominant in terms of our revenue percentage. As we get to 2030 to your question, we think it's going to be more balanced in a couple ways. Number one is some of the other segments that we talk about in the deck, including EV tools, including robotics. That is where we are spending some time today. Today, we've added to our sales force, as we mentioned, with European folks and others, and we've also stood up a separate business development arm so that we are... Practicing some of those further out opportunities. So that's something that we think about. I think the balance also from a supply side will be better balanced in three, four years. Customers are starting to ask for batteries from specific locations. And we are responding. That's part of why you see Kamenets. That's part of why you see what we're doing with our South Korean partners in the U.S. So that will come into the actual supply of that demand in that time frame and will be better balanced between China generally, Asia, mostly South Korea, and the U.S. at the end of the decade.
Operator
Thank you. Our next question comes from the line of Austin Bolling with Needham & Company. Please proceed with your question.
Hey, guys. Thanks for taking my question, and congrats on the solid results here. Maybe just wanted to first start on kind of the new updated guidance, third time you raised it. Would just love to maybe talk through kind of different scenarios and like maybe potential further upside to kind of this new target.
And, sorry, just to be clear, you're referring to the years or the longer-term one, Austin?
Yeah, so the 145, you know, we made it just to 145 simply because we run out of time here with the end of the year coming. And I do think that, you know, when it's time to put out the 2027 number, We've got pretty good visibility in that, but obviously the longer we can spend building that book, the more solid that will be as we communicate it. I do think that what is implied by the guide, which is around $43.3 million of revenue for Q4, It makes a lot of sense from where we stand, especially if we're trying to get to really strong gross margins. The guide, in essence, implies margins getting to almost 35%. And in order to do that, we really do need to be heavy on the pouch side. So that's the main goal. I do think that if we weren't so focused on managing the margins in Q4, to your point, there could be additional upside on the top line. But I do think that in this phase that we're in, demonstrating our path to profitability and the margin potential of the business, given that we have a pretty healthy backlog, Is more important than just continuing to build up the revenue base.
Okay. Very helpful. And I think, too, like, really it seems exciting opportunity in the subsea market. Just curious on how, like, the content per vehicle in that may defer to drones.
Yeah, good question. So we are also very excited about that. There's a lot of opportunity underneath the surface. We generally, these craft are larger. Compared to drones. So the battery content tends to scale with the overall size of the craft. Some of that depends a bit on the mission that the craft is actually doing. There are some one-way underwater drones. There's some surveillance drones also, just like we show in one of our diagrams. So, it's a new space for us. We are... Talking to many customers, and some of the same value that we provide, in particular, the ability to handle 9,000 feet, 3,000 meters, is somewhat unique to our design. So we want to exploit that and then try to get some more wins on the board.
All right. Perfect. Well, thank you, Tom, and keep up the good work, guys.
Operator
Thank you. Our next question comes from Ryan Finks with the O'Reilly Securities. Please proceed with your question.
Hey, good morning, guys. Thanks for taking the questions.
Hey, Ricardo, maybe just a follow-up on the revenue mix and thinking about 2027, understanding you're not giving guidance yet, but how should we be thinking about the potential breakdown next year between U.S. versus international opportunities as we try to size the potential for U.S. growth going into 27?
Yeah, the way we're thinking about it right now is that, you know, if the U.S. can double year over year, that would be a decent – that would be kind of its contribution to the mix for next year. A lot of it is supply-dependent as well, right? So the onus is on us to get the capacity at Kaminas in particular and in South Korea to be higher than what it is today. But, yeah, I think if we can do a doubling of our U.S. business going into next year, that would be sort of my best estimate for what the mix of it would be.
That's helpful. And then I wanted to touch on the progress of your South Korean. Contract manufacturing partners and the progress they've made growing their capacity, and maybe you could frame what that capacity looks like today or where you expect it to be by the end of next year.
Yeah, so at the August call, we gave the names of three of our contract manufacturing partners in South Korea. There's a fourth that we've added and officially signed up, and there's a little more that we're working with. Our team was over in South Korea two weeks ago. I'll be there in about 10 days myself, probably once again before the end of the calendar year. So it is a market we're spending an awful lot of time on to help stand them up. They are working well. What we're doing, and I think many folks understand this, is in order to protect our intellectual property, which is mostly centered around the anode, there is one CM in South Korea that is making our anodes for us, and we're providing coils to companies like Caminus that are putting together the batteries. So we have tried to be thoughtful about how we industrialize the various CMs in South Korea particularly, not just to get the scale and the cost that we want, but at the same time, make sure that we have protecting RIP and also making it faster for newer CMs like Kamenus to come up to speed. If we take a little bit off of their shoulders and they're focused just on assembly and test, for example, the back two-thirds, the middle, and the last part of a typical lithium-ion cell factory, focusing the first third at one trusted partner, that helps speed, it helps cost, it helps IP protection.
That's helpful detail. Congrats on the progress, guys. Ryan, thank you.
Operator
Thank you. Our next question comes from the line of Derek Soderbergh with Cantor Fitzgerald. Please proceed with your questions.
Yeah, hey, Tom. Hey, Ricardo. So, my first question is related to the earlier question on pricing, the $350 per kilowatt hour. Cycore pricing is higher than that, but curious, you know, how that is sort of trended, that figure, as volumes have ramped up in Korea. And, you know, I think, Ricardo, you had said that the $350 figure was conservative, but should we view that as more or less kind of a longer-term blended average price as you sort of get into the gigawatt scale?
Yeah, I think that, you know, the 350 is a very long-term floor that is very cylindrical heavy, right? So, if you look at what's on slide 6, I think it's probably fair to assume that, you know, 70% of that is cylindrical, which we obviously are trying to disprove as we create the market. And, you know, to your question on recent pricing, recent pricing has actually trended higher relative to what our expectations were, you know, going back just a little over a year ago, especially on some of the pouches. The cylindrical pricing is challenging. It's also more challenging to demonstrate better performance on a cylindrical cell, so that makes total sense, and that's why we're pretty incentivized to drive our pouch cell portfolio in every way we can.
Yeah, and Ricardo, can you touch on that a bit more? Just can you pull out the mix portion of the pricing increase you're talking about? What else has been contributing to that? Negotiations, better technology, is there anything else that's sort of driving pricing beyond mix right now?
Yeah, I think the main thing driving pricing right now is that there's no capacity out there from... You know, in the market, either you are stuck buying a standard lithium-ion cell, or you try to get a high-performance cell from a very short list of participants like us. On the cylindrical side, there is more competition, right? You know, the Molly cell has a... A high-performance sell in the cylindrical market that, you know, we think they're not pricing as high as they should, frankly, and there are several other players selling standard cylindrical cells at pretty low prices that sometimes make us even question, okay, you know, if we cannot source a cylindrical cell at a certain cost. Why do we bother with this, especially when we have so much demand on the pouch side? So that's something that we'll continue managing here as we build out the portfolio. And our team is not standing still, right? We're figuring out a way to get even more performance out of a cylindrical cell potentially.
Got it. Super helpful. That's all for me. Thanks, guys.
Operator
Thank you. Our next question comes from the line of Chip Moore with Roth Capital Partners. Please proceed with your question.
Hey, good morning. Thanks for taking the question. Maybe one on retrofit of the EV battery line. Just talk about progress on key milestones there and how those tie into future grant disbursements.
Yeah, Chip, thanks for your question. I'll take that. So the competitive contract, to use the appropriate language, that we've been coached from the Department of War, because it's exactly that, it's a competitive and a contract, has started. This time last week, actually, Wednesday last week, we had our kickoff meeting with the team in D.C., attended, of course, with our friends and partners in South Korea. There are 15 milestones. It's a 24-month program. The first part is about making sure we provide the correct tools that will go into the factory, and then there's some retrofitting of the factory itself that comprise most of our spending. We spend first, and then we are reimbursed for that spending. Some of that flows through Amprius, right? So we will work with our South Korean partners to spec up the tools. They will buy the tools. We will invoice the government. They will send us funds, and we will send most of those funds for tools in this example to South Korean. So that's how it's working. It's up and running. There's a lot of work to do. It'll start to look real once we start to deliver equipment to the factory and the retrofit. So that's probably summer, fall of 2027. But we're up and running. We're off to the races.
Excellent. Thanks, Tom. And maybe for my follow-up, you know, the slides around defense spending, slide five, with particularly a lot of, you know, European allied nations in there, and you also talked about some hiring over there. Just do you see a need for additional contract manufacturing over in Europe here sooner than later, or how are you thinking about that? Thanks.
We do. Some of the requests, I mentioned earlier that customers have specific requests. There's a law, of course, here in the U.S., the National Defense Authorization Act, that is very clear on what is required and when. The when is about 14 months from now, depending on the program, but generally the beginning of 2028. There is not a law in Europe generally, but there are some preferences that we are starting to hear from our customers. We have discussions underway. We were first... On a specific mission to look at some European CMs, oh gosh, a year or so ago, there's actually one of our engineers is in Europe visiting his folks, but he is stopping by and spending tomorrow with one of the CMs that we've seen in the past. As you can imagine, there's a lot of work that happens before we actually announce customers, right? We started working with Kamenus, oh, my gosh, probably in the spring. We announced it a month or so ago. So that will come. That will give us some of that same balance that I tried to answer. On one of the other questions. Thank you, Chip.
Operator
Thank you. Our next question comes from the line of Amit Dayal with H.C. Wainwright. Proceed with your question.
Thank you. Good morning, guys. Just one question from me. Hey, guys. So, you know, China recently implemented some tax policy changes that have been disruptive to the lithium ion industry over there a little bit, at least in the near term, you know, a lot of margin pressure, a lot of sort of orders being pushed out, etc. Does this development potentially help Amprius?
Potentially, right? As different countries are implementing either protective measures or other types of restrictions, we have to be nimble. And that's part of why we have more than one leg to stand on, South Korea, U.S., cylindrical pouch. So, we... Keep our finger on the pulse of what's happening around that front. We have a team. We have a 12-person team in China that is present. So we're on top of it. We have outsiders, government affairs firm, various lawyers who also keep us up to date on some of these geopolitical considerations, and that's part of That's part of how we run the show internally here.
So, Ahmed, the other thing to keep in mind is that a lot of this shakeout happens in, you know, way larger markets than what we're addressing, so mainly EVs and stationary storage, and so, you know, in some ways, I do think that we're sort of flying our plane either above or below that turbulence.
Understood. Thank you guys. That's all I have. Thanks. Thank you.
Operator
Thank you. Our next question comes from the line of Josh Sullivan with Jones Training. Please proceed with your question.
Just wanted to kind of get your thoughts on how Part 108 evolves. And, you know, whenever that happens, you know, how quickly is that going to translate to, you know, activity for you guys? You know, how do you think the market kind of evolves? I just wanted to kind of get your thoughts around that as we get closer.
Oh, that's the line of sight regulation, right?
Yeah, on the commercial side, BDLOS.
Yeah. Yeah, I mean, I think that'll unlock a good bit of non-defense demand, and there's a big ecosystem of companies that are going after that market, right? So, you know, drones as first responders seem to be the opportunity that is their most ripe to take advantage of this. We actually had an internal meeting here to... You know, cover our targeting of a lot of these names. Some of them we know well. And, yeah, I do think that in an, you know, kind of unclogged DC, if this actually goes into effect, that'll be a big tailwind for all of these participants.
Yeah, and again, different types of drones, right? So, DFR. As worried about VBLOS beyond visual line of sight, delivery, et cetera, we have some public names that we have talked about for those. And it's a land and expand strategy with a couple delivery companies under our belt, and we want to talk to more. So as that evolves and, frankly, makes it easier for our customers, we believe that we should be a beneficiary.
Great. I'll leave it there. Thank you. Thanks. Thank you.
Operator
Thank you. Our next question comes from the line of Ted Jackson with Northland Security. Please proceed with your question.
Thanks very much. Yeah, I wanted to talk a little – hey, guys. Yeah, you know, you guys saved the best for last, I'm sure. Totally, yeah. I wanted to talk a little bit about the North American market, you know, and where I'm going with it is it's just kind of a more of a discussion. But, you know, you had a big jump in it this last quarter in terms of its contribution. I'm assuming that was driven by, you know, kind of drone dominance. And I just wanted to verify that. And then when I think about kind of the business as we get into 27 and 28, there's a couple of things that cross my mind. One is you're going to ramp your North American production really in 27. It's going to be pouch-cell based. And so the first thing is, how does that change your mix? You had a high cylinder count this time, but will we see the bulk of your revenue going forward? Will we see the mix of business shift and be a little more North American because you have this capacity in place? And then because of it's a pouch sale and you're going to be making a lot of this stuff more in the North American market, how does that all play out with regards to the margin structure? For Amprius, because you're not going to have the freight costs, and you're going to be having, you know, basically your higher margin, you know, product line being, you know, produced here in the U.S. I guess that's more or less my only question to have.
Yeah, no, that's helpful. So, two points. I think, first, we shouldn't forget about the announcement that we made earlier this year around nanotech and our intent to have Cylindrical cell production here in the U.S. through that, and so we will have hopefully some, you know, good and competitive cylindrical cells here in the not-too-distant future to go after the U.S. market. We continue looking for, frankly, additional supply of low-cost, high-performance cylindrical cell. That would enable us to not completely tilt the mix towards pouches in that late 2027, 2028 timeframe, to your point, Ted. And so, but yeah, if the mix completely tilts towards pouches with, you know, pricing being the main driver, obviously. I do think we could potentially have an even more attractive gross margin profile, especially with the scale that we're bringing, right? So, I mean, if you look at the demand that we showed here on slide six, it was. And you put, you know, 300 or more megawatt hours of capacity, that's actually pretty good. And that, in many ways, explains the rationale behind the $75 million contract that we got with the DIU. So that is a gap in demand that we have to fulfill here in the U.S., and where we're pretty uniquely positioned to go in and do that if we can execute.
Yeah, maybe just a couple of things to add on there. So, as Emil Michael said, the U.S. is starting to get serious, right? He's imagining different scenarios, and we're starting to see that flow through, albeit there's uncertainty about the actual size of some of these budgets. And that's also evidenced by these gauntlets, right? There's four gauntlets on drone dominance. Two was... Number three is in February, I believe. So, the U.S. is starting to play catch-up. And then, by the way, just to be clear, we do have an $18 million defense innovation unit, and then the 75 is with a different group, the industrial base folks. So, both of those are underway.
Okay. Thanks very much, and congratulations on a superb quarter.
Thanks, Seth. Thank you. Thanks so much.
Operator
Thank you. Ladies and gentlemen, that concludes our question-and-answer session. I'll turn the floor back to Mr. Stepien for any final comments.
Thank you. To our customers, shareholders, employees, and partners, thank you for your continued support. Several markets need advanced batteries, and Ampria cells lead the industry on multiple dimensions, which puts us in a really strong position to keep innovating, execute with discipline, and deliver meaningful results. Thank you for your time and attention.
Operator
Thank you. This concludes today's conference call. You may disconnect your lines at this time. Thank you for your participation.
