VICR - NASDAQ

Good day, and thank you for standing by. Welcome to the Vicor Fourth Quarter 2025 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your first speaker today, Jim Schmidt, Chief Financial Officer. Please go ahead.

Thank you. Good afternoon, and welcome to Vicor Corporation's Earnings Call for the Fourth Quarter and Year ended December 31, 2025. I'm Jim Schmidt, Chief Financial Officer. And I'm in Andover with Patrizio Vinciarelli, Chief Executive Officer; Phil Davies, Vice President, Global Sales and Marketing. After the market closed today, we issued a press release summarizing the financial results for the 3 months and year ending December 31. This press release has been posted on the Investor Relations page of our website, www.vicorpower.com. We also filed a Form 8-K today relating to the issuance of this press release. I remind listeners, this conference call is being recorded and is the copyrighted property of Vicor Corporation. I also remind you various remarks we make during this call may constitute forward-looking statements for purposes of the safe harbor provisions under the Private Securities Litigation Reform Act of 1995. Except for historical information contained in this call, the matters discussed on this call, including any statements regarding current and planned products, current and potential customers, potential market opportunities, expected events and announcements and our capacity expansion as well as management's expectations for sales growth, spending and profitability are forward-looking statements involving risks and uncertainties. In light of these risks and uncertainties, we can offer no assurance that any forward-looking statement will, in fact, prove to be correct. Actual results may differ materially from those explicitly set forth in or implied by any of our remarks today. The risks and uncertainties we face are discussed in Item 1A of our 2024 Form 10-K, which we filed with the SEC on March 3, 2025. This document is available via the EDGAR system on the SEC's website. Please note the information provided during this conference call is accurate only as of today, Thursday, February 19, 2026. Vicor undertakes no obligation to update any statements, including forward-looking statements made during this call, and you should not rely upon such statements after the conclusion of this call. A webcast replay of today's call will be available shortly on the Investor Relations page of our website. I'll now turn to a review of our Q4 and full year financial performance, after which Phil will review recent market developments and Patrizio, Phil and I will take your questions. In my remarks, I will focus mostly on the sequential quarterly change for P&L and balance sheet items as well as full year-on-year changes and refer you to our press release or our upcoming Form 10-K for additional information. As stated in today's press release, Vicor reported product revenue for the fourth quarter of $92.7 million, up 4.5% from the third quarter total of $88.7 million and up 15.3% from the fourth quarter 2024 total of $80.4 million. Royalty revenue for the fourth quarter totaled $14.5 million, a 33.1% sequential decrease from $21.7 million in the third quarter and a 7.8% decrease from $15.8 million in the fourth quarter of 2024. The sequential decrease in royalty revenue was the result of a catch-up amount that was included in the Q3 results. Product revenues for the year ended December 31, 2025, increased 12.1% to $350.3 million from $312.5 million for the prior year. Royalty revenue for the year ended December 31, 2025, totaled $57.4 million, a 23.2% increase from $46.6 million for the year ended December 31, 2024. Total product revenue and royalty revenue, including a $45 million patent litigation settlement received for the year ended December 31, 2025, increased 26.1% to $452.7 million from $359.1 million for the prior year. Advanced Product revenue, which includes royalty revenue, decreased 4.4% sequentially, which was the result of the catch-up amount of royalty revenue in Q3. Brick Products revenue declined 0.6% in the third quarter. Revenues for Advanced Products for the year ending 2025 increased 26% to $248.6 million from $197.3 million the year before. Revenues for Brick Products for the year ending 2025 decreased 1.6% to $159.1 million from $161.7 million the year before. Shipments to stocking distributors decreased 11.1% sequentially, but increased 5.3% year-over-year. Exports for the fourth quarter increased sequentially as a percentage of total revenue to approximately 49.3% from the prior quarter of 42.8%. On a year-over-year basis, exports increased as a percentage of total revenue to approximately 50.8% from the prior year's 48.2%. For Q4, Advanced Products share of total revenue, including royalty revenue, decreased to 58.4% compared to 59.4% for the third quarter, with product share correspondingly increasing to 48.6% of total revenue. Turning to Q4 gross margin. We recorded a consolidated gross profit margin of 55.4%, approximately 2.1% less than the prior quarter as a result of the royalty catch-up in Q3. For the full year 2025, gross margin rose by 6.1% to 57.3% from 51.2% in the prior year. I'll now turn to Q4 operating expenses. Total operating expense increased 2.7% from the third quarter. For the full year 2025, total operating expense as a percent of revenue and patent litigation settlement decreased to 39.2% from 51.6% in the prior year. The amounts of total equity-based compensation expense for Q4 included in cost of goods, SG&A and R&D was $1.08 million, $2.206 million and $1.153 million, respectively, totaling approximately $4.4 million. For Q4, we recorded operating income of $15.7 million, representing an operating margin of 14.6%. For the full year 2025, operating income totaled $81.8 million or 18.1% of revenue in patent litigation settlement compared to operating loss of $1.3 million or minus 0.4% of revenue in the prior year. Turning to income taxes. We recorded a tax benefit in Q4 of approximately $27.3 million, representing an effective tax rate for the quarter of minus 142% as a result of the tax benefit due to the partial recognition of certain deferred tax assets in the period. The tax benefit for the full year 2025 was approximately $24 million, representing an effective tax rate for the year of minus 25.4%. Net income for Q3 totaled $46.5 million GAAP diluted earnings per share was $1.01 based on a fully diluted share count of 46,297,000. For the full year 2025, net income increased to $118.6 million from $6.1 million in the prior year. In 2025, fully diluted earnings per share increased to $2.61 from $0.14 in the prior year. Turning to our cash flow and balance sheet. Cash and cash equivalents totaled $402.8 million in Q4. Accounts receivable net of reserves totaled $60.7 million at quarter end, with DSOs for trade receivables at 44 days. Inventories net of reserves increased 1% sequentially to $91.3 million. Annualized inventory turns were approximately flat sequentially at 1.96. Operating cash flow totaled approximately $15.7 million for the quarter. Capital expenditures for Q4 totaled $5.5 million. We ended the quarter with a construction in progress balance primarily for manufacturing equipment of approximately $7.8 million, with approximately $6.9 million remaining to be spent. I'll now address bookings and backlog. Q4 book-to-bill improving sequentially, came in well above 1 and with 1-year backlog increasing 15.8% from the prior quarter, closing at $176.9 million. 2026 is a year of great opportunity for Vicor. We are working to deliver on the opportunities. However, given that we cannot predict with certainty the timing or amounts of outcomes relating to our licensing practice, we will not provide quarterly guidance. With that, Phil will provide an overview of recent market developments. And then Patrizio, Phil and I will take your questions. I ask that you limit yourself to one question and a related follow-up so that we can respond to as many as we can in the limited time. If you have more than one topic to address, please get back in the queue. Phil?

Thank you, Jim. At the beginning of 2025, we talked about the year ahead being one of challenges and opportunities. As we look back, 2025 met those expectations with improvements in product bookings and revenues in Q4 and our IP licensing practice becoming a major contributor to our top and bottom lines. As we exited 2025, book-to-bill ratio increased to over 1.2 in Q4 and has continued to increase in Q1. At the start of 2026, we can say that this will be a year of different challenges and greater opportunities. This should result in record bookings, revenues and profitability and significantly higher utilization of our first chip fab. As Patrizio commented in today's press release, United States International Trade Commission has instituted a second investigation into illegal importation of power modules and computing systems, infringing Vicor's IP to nonisolated bus converters. By now, it should be clear that Vicor will methodically and relentlessly enforce its intellectual property to the many inventions it pioneered and that suppliers of infringing systems are putting themselves and their customers at risk, including unlicensed OEMs and hyperscalers. Following the example set by licensed OEMs and hyperscalers, companies with an ethical backbone should do the right thing, avoiding infringement by taking a license to secure their supply chain. Our lead customer for VPD solutions is ramping a Gen 4 factorized power system before transitioning to a Gen 5-based solution with higher current density and performance. This transition is expected to start in the second half of this year, while production of the Gen 4 system will continue to ramp at a steep rate to the end of 2026. Engagement with other Gen 5 VPD customers will be selective as capacity in our existing first chip fab is getting earmarked for strategic customers and additional capacity from our second chip fab may not be available until 2028. Our industrial and aerospace and defense business outlook for 2026 is strong, particularly in the automatic test equipment market, which is seeing substantial growth and projecting high growth to last for the next several years. Given our power density advantage, which is of paramount importance to our customers, I am confident that we can double the revenues in these markets over the next 4 to 6 years, respectively. As we approach high utilization of our first chip fab, we are beginning to engage customers in capacity reservation agreements to secure their supply needs. While in the planning stages of a second chip fab to expand the market opportunity, we are having discussions with candidates for an alternate source of high current density Gen 5 VPD solution. An alternate source will give licensed OEMs and hyperscalers broader access to best-in-class power system technology. In view of these developments, we remain confident in our business strategy of innovation, customer focus and market focus. With that, we'll now take your questions.

Our first question comes from the line of Quinn Bolton of Needham & Company.

Congratulations on 2025 and the record outlook for 2026. Patrizio or Phil, I wanted to start with your lead customer. It sounds like you're seeing a pretty strong ramp from that customer. And you mentioned that Andover is getting filled. Can you talk is Andover being filled largely from your lead customer? Or do you have other significant Gen 4, Gen 5 customers that are contributing to that growing utilization in the Andover facility?

It's a combination of demand, increasing demand on a number of fronts, not just hand computing where there is a multiplicity of factors at play with respect to increased demand on capacity, but also in test equipment, as Phil mentioned in his prepared remarks and some of the other end markets.

Got it. Okay. And then I guess maybe a follow-up on the IP licensing. In the press release, you talked about seeing record revenue from the IP licensing business this year. Just wanted to clarify, does that include or exclude the $45 million patent litigation settlement that was part of the 2025 revenue stream as we think about 2026?

We see our licensing business expanding. As Jim suggested earlier, the timing of elements contributing to the expansion is somewhat unpredictable. But as we look at the predicament that OEMs and hyperscalers face in terms of potential exclusion orders, we see a major opportunity for us to grow our licensing business considerably. As we have discussed in our last quarterly call, we see that business expanding greatly in the last couple of years. I think what has transpired since then suggests that those are conservative estimates.

And Quinn, just to clarify the number for you, the royalty revenue I quoted in my prepared remarks of $57.4 million in 2025 does not include patent litigation settlement. That's royalty revenue, it was up 23.2% from $46.6 million in 2020.

Got it. But just to clarify, Jim, when the comment in the press release about the business, the licensing business will expand, are you looking at the $57.4 million as the 2025 base? Or should we be thinking about that base being $102 million, which would include that $45 million patent settlement as part of the base?

So for one thing, there's going to be more patent settlements. And for another, the one patent settlement from last year in terms of the outlook for licensing business, it doesn't really make a substantial difference with respect to the upside with respect to this part of our business. We expect hundreds of millions of dollars worth of revenues from licensing and the $47 million event last year is in hindsight going to be in the bucket to be sure, but not all that significant.

Our next question comes from the line of John Tanwanteng of CJS Securities.

And also congratulations on a good year. I was wondering if you could give us a little bit more detail on the launch customer for VPT. You mentioned that they were going with the Gen 4 product. Could you talk about the decision that went into that and why they aren't starting with Gen 5 and kind of how that happened?

Well, so the Gen 4 system is mature. It's one that's got a track record of success that is expanding in terms of its opportunity in order to get to the next-generation system, mature design, a mature system. It's not just the power systems, the system as a whole needs to come to fruition. It isn't quite there yet. It will be there soon, and that will lead to the next set of opportunities. But to be clear, with our lead customer, we're seeing a significant share of our capacity being utilized as we get towards the end of this year on the earlier generation system. And the next-generation system will provide an additional layer of use of capacity as we get into next year.

Understood. And then when you start -- sorry, you're considering a new facility. I was just wondering if you're planning to build that yourself or are you still planning to work with partners to do that perhaps in a capital-light fashion. And just wondering what kind of capacity a new facility would have?

So we've made 2 offers on area where we could build a facility. The lead time associated with that though is 1.5 to 2 years when everything is done. We are also looking at existing buildings within 30-mile radius of and over to the North and the West. And we haven't decided yet which of these alternatives we're going to close on. But again, we've had 2 offers. No deal done yet, but I would expect that we're likely to do something on this soon.

And our next question comes from the line of Richard Shannon of Craig-Hallum Capital Group.

I'll also add my congratulations on a really good last year. My first question is on royalties and licensing here. As you mentioned, there are some questions here in the Q&A about growth in this business. I guess I wanted to triangulate it differently from how you've talked about in the past where you're hoping to get a roughly $300 million revenue stream. I know that's not entirely royalties, maybe some product in there, but talking about $300 million bogey between '24 and '26. And by my numbers, at least that would require a fair amount of growth, like doubling or so of your royalty revenues from '25 to '26. But you didn't talk about it that way this quarter. Can you maybe talk about it in those terms here? Is that a number that we should continue to expect better or worse? Just help us triangulate those things.

Yes. So we have 2 major licensees. We expect to have a lot more and future contribution from those 2 should become quite a bit larger. So I think in one way of looking at it, in high-end computing AI, systems require from the power system perspective, our IP. And to the extent that in order to be able to deploy those systems, a license will become necessary. that defines the opportunity. As you can see, the opportunity far exceeds what we've harnessed thus far. There's a lot more to be captured in years to come. So the $300 million number, which to your point involves contributions from royalties and business, licensees is not a long-term goal. It is a relatively near-term goal, not for this year, to be clear, but as we have said last year in a couple of years' time frame, but we can see going beyond that.

My follow-on question is on second-gen VPD engagements. You already talked about your lead customer today and in past quarters. But last quarter, you also mentioned engagements that didn't seem to be early stage ones with a hyperscaler and an OEM. And I didn't hear any comments in the prepared remarks, although I was a little bit late. So wondering if you can comment on the progress of those and any other ones you've added to the pipeline.

Yes. So Richard, this is Phil. So maybe I can get a little bit more granular on that. So the next step for us is over the next couple of weeks, we're bringing in our global FAE team that is dedicated to supporting customers in different locations where we have target hyperscalers and OEM chip companies located. So they will be going through, if you like, a boot camp on Gen 5 VPD using the demo boards and tools that central applications group here Andover have developed for the market. And so that's happening in the next couple of weeks. After we get that in place, as we talked about, we're going to be fairly selective in who we're going to be engaging with. It's very important we do that. And so that's the next step after that. So we FAEs are here in the next couple of weeks, and we're on the way.

Our next question comes from the line of Justin Clare of ROTH Capital Partners.

So first, I just wanted to follow up on the potential for capacity expansion here. So given the plan to add a second fab, I was just wondering if you -- how we should think about the ramp in utilization for your existing facility, how we think about that over the next couple of years and kind of what utilization threshold you anticipate reaching that is necessitating the additional fab here? And then just if you could talk about when do you anticipate kind of approaching that optimal utilization for the first fab?

So based on ramps with our customers in different markets with a strong contribution from high-end computing, we see the existing fab being well utilized within a year. And that's obviously prompting the initiative to secure additional capacity, both by bringing up a second fab and by having discussions with potential alternate sources that could provide customers with equivalent solutions using their own capabilities and our technology. In terms of the fabs, as I mentioned earlier, we've started exploring the opportunity of being about with a large piece of real estate, the flexibility to increment capacity in steps. As I remind you, this will be a campus that could support up to 0.5 million square feet of manufacturing space. Just to set things in perspective, the facility is around 300,000. So there would be substantially more in terms of the real estate available for capacity. But also given the learning that we've done, we think we can achieve more capacity per unit of area in our next facility. The thinking of late has evolved though more towards potentially acquiring a building. And to be clear, there's been no decision one way or the other yet. It could go either way. But the benefit of doing it with an existing building is that the time to fruition would be year, 1.5 years shorter. So we might go that way. It would be on the same scale, though, in terms of the increment of capacity that we want to bring about with the second.

Okay. Got it. That's helpful. And then just when you think through this, if you're reaching close to kind of optimal utilization within a year, I think historically, you've talked about your fab being able to support $1 billion in product revenue. So within a year, could you be close to that level where you're getting to a run rate of $1 billion in product revenue? And then just curious on the second fab, how much in CapEx spending you might anticipate in terms of what's required there?

Yes. So to be clear, fleet has a capacity, given the dollars per panel and the number of panels it can process of time to do slightly above $1 billion in revenues. But you wouldn't want to use 100% of the capacity because by definition, that will leave no room for error, right? So an 80% capacity utilization is the kind of number that you want to think of in terms of the test or fundamentally having achieved a very good capacity utilization. Now in terms of the next facility, whether it's by acquiring land, putting up a building and then equipping it with what is necessary in order to bring about that relative incremental capacity. This is in a proposition of the order of $250 million, $300 million, something that Vicor has to finance on its own given our cash position and our balance sheet.

And our next question comes from the line of [ John Dillon of DMB Capital ].

Again, congratulations on a good year. Phil, I wanted to go back to the customers you talked about before in Q3 and Q4. I kind of got the impression that you had design wins and these customers couldn't find alternative ways to power their new AI processors. So I'm wondering, are those customers still working with you? Or they -- have they gone to other customers? Are you going to be able to meet their time schedule for their new products?

They have a need for a VPD solution in particular that has all of the right ways. And the competitive landscape doesn't have that. And that's constrained the market opportunity for VPD to a very limited set of companies that have actually done it while incurring a great deal of pain because of the shortcomings of the power system. So what we bring about with our second-gen VPD and fifth generation modules is a solution that has much higher density and much higher greater level of manufacturing quality in terms of the assembly of the solution. It doesn't require stack as generation VPD does. It's much easier to pool. It's more efficient. It has a number of benefits that manifest themselves in many ways. So as Phil suggested, we're going to be picking those customers that strategically want to be aligned with. We have a great deal of interest. As an example, we were out in the valley just a few weeks ago, in the morning with an automotive customer with a great deal of interest in our VPD capability. We haven't decided yet whether or not we're going to engage in that particular case. We will be in a situation with our existing fab before we get another fab in place of deciding which applications make the most sense. And to say a lead customer is one that we prioritize. There's going to be more in that league in that end market. There's one in particular with tremendous opportunity in terms of volume. That one alone fill 2 fabs. So we are in a privileged position. We have the technology and the capability. We can leverage our opportunity both by selling product and by collecting licensing. We can also do it by bringing about we're pursuing all these opportunities evolve.

Got it. So I just want to make sure I understand. So the customers that you mentioned before, they're still on the hook. They're still talking to you. They're still engaged with you. They still can't find an alternative source to power their new AI processors, but it sounds like it just slipped a bit.

Well, I think if you were to ask them, they would all say that they will find a solution, but not the driver exists, right? Nobody will acknowledge that they're out of luck without us. And that's not the real world. That's not what we're suggesting. There's always some way of getting something done. But to be clear, that way of getting it done is problematic in terms of the technical trade-offs and technical challenges, whether it's cooling or manufacturability. And then it may also be very much challenged from the IP perspective. So it's a complex landscape, but...

It sounds like it's still a competitive situation then.

Well, it's always beneficial competitive situation.

Yes. Got it. So my follow-on question is, are you seeing any AI processor designs with horizontal or horizontal vertical besides your lead customer?

So I think if you look at -- as Patrizio actually said, there's one very, very large company that's using vertical power delivery today in very high volume, and that's increasing year-on-year. In terms of anybody else really in high-volume production, it's with vertical power delivery, [ John ], it's fairly limited right now. They're all trying to get Gen 1 VPD to work in some fashion. But to date, I'm not hearing anybody that's deploying that in volume. They're trying. They're working on it. But I think when we come out with our Gen 5 and launch it and selectively launch it, as we've talked about, we're going to have some winners on our hands.

Got it. I saw a picture of a new AI processors coming out that had -- it looked like a gold bar on the top. And that's why I ask about horizontal. I'm wondering if you have any upcoming horizontals or horizontal verticals besides your lead customer because I know they're different.

So I don't think we're going to make comments specifically about that stuff. I think what do you want to say about that?

We do have Gen 4 customers using our gold bars collaterally...

But I don't think the visibility to a gold bar is really what's fundamentally add issue at this point. I think the right way of looking at it is that we have tremendous opportunity, and we have the technology that matches the needs of the marketplace. Again, going back to the earlier question, it's not that if our solution didn't exist, there wouldn't be a solution. But the alternative solution, which is really a common to all the competitors that tend to do pretty much the same thing with relatively slight differences as they look over each other's shoulder to make incremental steps down and over the road. It carries a lot of baggage in a number of respects, technical and when it comes to APD, also IP challenges.

And our next question comes from the line of Jon Tanwanteng of CJS Securities.

Earlier, you mentioned that you were taking capacity reservations for your facility. I was wondering what the financials of that look like? Is there an upfront payment? Are there contract terms for minimums or something like that? Just how are you approaching that -- those reservations?

So in terms of revenue recognition, that would happen as shipments take place. Obviously, there's a cash component that would show up in our balance sheet. But there is no acceleration of revenue that comes from the capacity reservation. The revenues get recorded as products ship covered by that reservation.

Okay. Got it. And then can you talk a little bit more about the 800-volt data center opportunity? And if you are seeing any traction there? Or are you seeing any orders ahead of that? And I'm specifically talking about products that are outside the vertical lateral power or the NBMs that you have today?

So we have technology there. there too, Lagos pioneered high-density past conversion from 800 volt and 400 volts for many, many years. We have relevant IP. We have products. We have more products in the pipeline that will come out later this year. Frankly, though, I would say that there is quite a bit of hype about this 800-volt. I think that it's, to some degree, missing the point with respect to what the real issues are. It's a diversion. The reason why generations of GPUs have not been able to meet the expectations with respect to performance having to do with the power system gating the GPU performance not to do with 48 volt or 800 volt they had to do with what goes on at the point of load and the fact that multiphase mainstream type of solutions are handicap. That's where the funnel should be. So obviously, we operating an industry that goes through phases of focus and progress and potential life. Without question, there is value to 800-volt bus. But that value probably if you measure in terms of efficiency, gets measured in a few percent. What gets lost in an inferior type point of solution is 15 or 20 points. So I personally wonder why anybody would worry about capturing a 3% improvement in 100-volt power distribution when they're missing 15% or 20% in the point of load and they can't call or deliver the power they need in order to achieve the level of performance they targeted. But irrespective of how these things evolve, we have the technology, we got the IP, and we're going to make the most of the opportunity. But frankly, I think there's going to be a lot of hype relating to as. And that could lead to problems because if people are focused on the wrong problem, which is not really mature problem, they are going to be solved in the real problems.

Our next question comes from the line of Quinn Bolton of Needham & Company.

Patrizio, I guess I just wanted to sort of make sure everybody on the line is sort of thinking about the revenue ramp the same way. You haven't obviously guided revenue for '26, but you've given us sort of 3 kind of guideposts, which are you expect Andover to become or to approach full utilization over the next year. You sort of said full utilization would be around 80%. Otherwise, you don't leave a lot of room for error. -- you said at 100% utilization, the fab would be able to produce $1 billion in revenue. And so when I put all of that together, it sort of sounds like you're pointing revenue could approach an $800 million product revenue could approach an $800 million run rate over the next year, and that would be more than double what you did on a product revenue front in calendar '25. I know you're not giving guidance, but some of those guideposts point to very significant revenue growth. And I just want to make sure to the extent that you think that interpretation of the comments you've made is too aggressive. I just wanted to see if you would correct any of those thoughts or if that's the right way to be thinking about sort of the data points you've suggested.

I think your analysis is on point. Obviously, key to that is run rate, as you think from revenues for this year, '26. So we see the demand getting to a run rate that would utilize 80% or so of the capacity in facility. Another way of taking this is that we see this year as being one of major increase in product revenue, well above the rate of last year and at a level that we haven't enjoyed for quite some time. And that's pretty much baked in at this point based on bookings that we received and additional bookings we expect to come our way as the year progresses.

And our next question comes from the line of Richard Shannon of Craig-Hallum Capital Group.

Let me ask a couple of follow-on questions here. My first one is on licensing here. Patrizio, following up on an answer to one of the prior questions here, you mentioned about having a couple or specifically 2 licensees so far. As we think about growing the licensing revenue stream this year, and if you can comment beyond that, that would be great in terms of kind of your general expectations. But how do we think about adding to the customer list here versus number of licensees or licenses per licensee or other dynamics that help us think about this? And I guess, specifically, if you could address if things went well for you, what's the kind of number of major licensees would you have? I don't know if this is 3 or 5 or 8, but if you can just characterize that in any way, that would be helpful.

In the high-end computing AI market, I think in terms of substantial licensees, it would be half a dozen. So 3x as many as we currently have in that market. minor ones on top of that. And by the way, the focus has been and the actions of the ITC thus far have been focused on high-end computing, but there's infusion going on in other markets as well. So there's a lot of opportunity, not just for the NBM technology, but for other technology that pioneer...

Okay. My follow-on question is wondering if there's any way that you can help us think about -- specifically about your second-gen VPD technology, how do we think about content per XPU? And I'm going to offer a couple of ways maybe to think about this. I know you're not going to quantify in any specific way, but I think a lot of us who cover this name for a while have a decent idea of what that content looked like a few years ago in your last really high volume or potential high-volume win that you had in point of load. But also since that time, the level of power and the level of current in leading XPUs, particularly getting to reticle limit, are increasing a lot here. So do we think about the kind of the content opportunity now as kind of being proportional to power current? And how do we think -- how would you have somebody think about what that might look like on a per unit basis?

So as I look back at a power system for GPUs a number of years ago, that was in one way of looking at it, about $100 million per year type of opportunity rising. We are locking into an opportunity that will double that. And to Phil's earlier point, there is a hyperscaler with an opportunity that could be another. I don't know if that answers your question.

Mine was really more on content per XPU, but the way you characterized it is also helpful. But any ways you might think about it on a per XPU basis would be helpful, too.

Yes. So Phil, do you want to take that?

Yes. I think, Richard, to your point, it really depends on the current that XPU, the number of rails, that type of thing. So I think that the opportunity for us will be somewhere between $200 to $400 per XPU. But very much depends on, right? So...

Take that with a grain of salt.

Understood. That's getting us a half order magnitude is very helpful.

So Richard, just clarify, it's about like a 2,000 amp up to a 4,000 amp type of product.

Our next question comes from the line of A. Hicks of Ainsley Capital Management.

I just wanted to confirm it's $1 billion capacity now.

I think we lost part of your question. I think the question was you wanted confirmation of the $1 billion capacity of Fab 1. Was that the question?

Yes. Yes, just for Advanced Products, nothing else.

Yes. We are very confident that we can generate upwards of $1 billion worth of revenues out of.

Okay. Because I'm looking at what your sales were for just for Advanced Products, not -- without royalties for the year was around $200 million. Is that for 2025?

Yes.

Okay. So you're saying within a year or so, you could be at $800 million in Advanced Products?

As suggested in an earlier question and confirmed by me, that would be a run rate.

Okay. And then on the bricks, the original bricks fab, could that be converted in the future to Advanced Products?

So no, the bricks much older products. They've got a very stable, if you like, customer base. So some of those customers are moving to advanced products, and we've had quite a bit of success of that in recent years in some higher volume end markets, but aerospace and defense and some very broad-based industrial, they like the bricks. They're going to stay with the bricks. So the brick piece will be fairly stable over the next few years. I believe...

Bricks don't really play a role with respect to capacity utilization. They become -- percent business, they become essentially.

Okay. But you're also adding capacity to this first fab. Is that correct also?

Yes, we are. Yes. So that's right. We're adding capacity incrementally to the existing footprint.

Okay. And then did you say you're in discussions with a partner to have them produce products themselves?

Yes. So we are having discussions. So we -- this may take some time because it's an important decision selection. But we have customers that want us to have an Altair source. We see the benefit of an Altair source in terms of expanding the market opportunity. If you just look at AI, there is so much of a market opportunity that frankly, there is no way that Vicor alone could do it even with the second and third fab. So we need to, in effect, look at making the most out of the opportunity as opposed to limiting the scope of the opportunity by wanting to do it.

Then I was just kind of curious, how many panels can you produce in a day out of the factory you have now?

I'm not going to quantify that for competitive reasons. I will just say that in terms of the revenue opportunity of the fab, Fab 1 is slightly above $1 billion a year.

Our next question comes from the line of [ John Dillon of DMB Capital ].

I'll make this quick because I know we're up against the time line. First of all, Patrizio, thank you for answering Quinn's question. That was one of my follow-up questions also, and I appreciate that answer. My another one is just a quick one. We're halfway through the quarter, and I'm just wondering how bookings are looking so far this quarter.

I mentioned in my prepared remarks, [ John ], that the book-to-bill was 1.2 in Q4, and we're above that already in Q1.