Lam Research Corp (LRCX) — Q4 2025 Earnings Call Transcript

Thank you, and good afternoon, everyone. Welcome to the Lam Research quarterly earnings conference call. With me today are Tim Archer, President and Chief Executive Officer; and Doug Bettinger, Executive Vice President and Chief Financial Officer. During today's call, we will share our overview on the business environment and we'll review our financial results for the June 2025 quarter and our outlook for the September 2025 quarter. The press release detailing our financial results was distributed a little after 1:00 p.m. Pacific Time. The release can also be found on the Investor Relations section of the company's website, along with the presentation slides that accompany today's call. Today's presentation and Q&A include forward-looking statements that are subject to risks and uncertainties reflected in the risk factors disclosed in our SEC public filings. Please see accompanying slides in the presentation for additional information. Today's discussion of our financial results will be presented on a non-GAAP financial basis, unless otherwise specified. A detailed reconciliation between GAAP and non-GAAP results can be found in the accompanying slides in the presentation. This call is scheduled to last until 3:00 p.m. Pacific Time. A replay of this call will be made available later this afternoon on our website. And with that, I'll hand the call over to Tim.

Thanks, Ram, and good afternoon, everyone. Lam delivered another great quarter. Revenues and profitability came in at the upper end of our guided ranges. Our gross margins exceeded 50% for the first time since the merger of Lam and Novellus, and EPS hit a new high for the company. We achieved record foundry revenues driven by strong performance in both gate-all-around and mature node markets. Our upgrades business grew to a new high, up mid-teens percent over the prior quarter as NAND customers migrate to higher layer count, higher performance devices to meet the faster read/write requirements and greater storage demands of AI applications. In short, we are executing well on the served market expansion and share growth story we laid out at our Investor Day earlier this year. 3D scaling of device and advanced packaging architectures is accelerating growth in etch and deposition intensity, and Lam's new products targeting key technology inflections are winning with customers. Furthermore, in advanced services, Lam is at the forefront of realizing the vision of an autonomous fab. We are gaining momentum with our Equipment Intelligence enabled Dextro cobots, which provide device makers with an unprecedented level of equipment maintenance precision and repeatability. The result is enhanced tool-to-tool matching, improved machine availability, lower operational costs, and, in some cases, higher yield. In the June quarter, Lam expanded our Dextro capabilities to cover three additional tool types, and we are accelerating the roadmap to support more products and more shipments in the coming quarters. Turning to the overall market environment for calendar year 2025. We expect wafer fabrication equipment, or WFE spending to be in the $105 billion range, up from our prior view of approximately $100 billion, predominantly due to an uptick in domestic China related spending. We see non-China investments remaining broadly consistent with our prior view. Currently, we expect WFE in the second half of the calendar year to be roughly flat with the first half. Looking forward to 2026, it is still too early to comment on the overall level of WFE spending. However, with our strong position in gate-all-around, advanced packaging, high-bandwidth memory and NAND layer conversions, we feel confident that Lam is well positioned to outperform. In 2025, Lam's served available market, or SAM, is set to expand to around mid-30s percent of WFE due to these industry drivers, and we expect them to work in Lam's favor again in 2026. Longer term, we are on a solid path to grow our SAM to the high 30% range of WFE by continuing to deliver critical solutions for atomic level device scaling, new materials innovation, and advanced packaging integration. The key R&D investments that we've made over the last several years have enabled us to create the broadest, most competitive product portfolio in the company's history, thereby putting us in a strong position to win over 50% share of the incremental SAM over time. I'll share a couple of examples that underscore our early progress towards our SAM expansion and share gain goals. First is our Halo ALD Moly tool, which is ramping at multiple NAND customers this year. We expect moly adoption to increase broadly as more customers convert NAND capacity to 200 layers and higher in the next few years. Lam is leading the industry transition to ALD Moly not only in NAND but also in foundry/logic. AI is driving greater transistor performance requirements and, in turn, accelerating the inflection to gate-all-around device architectures. However, below 2 nanometers, gate-all-around structures begin to encounter significant resistance capacitance or RC challenges. Narrower transistor contacts in these devices cause greater electron gathering, resulting in higher resistance of the deposited tungsten films. Replacing tungsten with molybdenum solves the resistance problem, but the process of depositing this higher performance material is inherently slower and more complex. This is driving a roughly 3x increase in Lam's metal deposition SAM per wafer when transitioning to advanced gate-all-around nodes. Today, we are the only company with ALD Moly tools already in production in foundry/logic. In the June quarter, we secured a key win at another leading foundry customer for their next-generation application. As moly adoption expands across various metal interconnect layers, the flexibility of Lam's unique multi-station architecture to execute both plasma and thermal processing in the same chamber enables optimization of process conditions and process step sequencing to meet requirements for different applications and over multiple generations of future logic devices. Another area where we have made strong progress is advanced packaging. Advanced packaging is critical for scaling system performance to address next-generation AI requirements and has enabled up to 100% improvement in memory density, 4x improvement in bandwidth and an approximate 40% gain in power efficiency. Lam's SAM is growing with greater adoption of next-generation packaging architectures for DRAM, CPUs, GPUs, and ASICs used in data centers. In 2021, leading-edge foundry/logic customers spent just 1% of WFE on advanced packaging. With AI's rapid adoption of advanced packaging, that number has grown more than 6x. In the future, we expect end consumer devices like mobile application processors and laptop CPUs to also feature more complex packaging schemes as on-device AI becomes mainstream. We are a leader in the advanced packaging inflection and are leveraging our experience to win more opportunities. For example, Lam has unmatched experience in copper plating hardware design and process technology over the last 20 years. We have, by far, the largest installed base in the industry and recently achieved a significant milestone of 6,000 installed plating sales. By incorporating our learning from the installed base into improvements in our latest SABRE 3D system, we are delivering best-in-class collinearity, uniformity, and defectivity in high-volume advanced packaging environments. The experience we have gained at the leading edge is now cascading to additional wins with next-tier customers seeking to adopt a proven best-in-class solution. SABRE 3D market share in advanced packaging is expected to grow nearly 5 points year-on-year in calendar 2025. Finally, let me pivot to the strong momentum we're seeing with our newest generation etch tools. In NAND, we continue to solidify our leadership in high aspect ratio dielectric etch. Equipped with our cryo process, our Vantex system recently won a key multigeneration etch decision at a major NAND customer. This further confirms our differentiation in both technology innovation and high-volume production worthiness in the NAND segment. Across all device types, our state-of-the-art conductor etch tool Akara is off to a solid start since its launch earlier this year. By combining direct power coupling with Lam's unique plasma pulsing capabilities, Akara delivers industry-leading depth uniformity and profile control that is vital for DRAM scaling. In the June quarter, Akara secured multiple new application wins at a top DRAM maker. So to wrap up, I'm excited about the breadth of opportunities I see ahead for the company and encouraged by the outstanding progress our team has already made towards the long-term goals we communicated at our February Investor Day. Etch and deposition intensity is rising with 3D scaling. Our products are winning in key technology inflections. And as a result, there is tremendous potential for Lam to continue expanding SAM and to grow share at each successive process technology node. Now here's Doug to talk about our quarterly financial performance and the September outlook.

Excellent. Thank you, Tim. Good afternoon, everyone, and thank you for joining our call today during what I know is a busy earnings season. We executed well on the June 2025 quarter, including delivering a record gross margin percentage of 50.3% in the post-Novellus period. These past two quarters represent Lam's highest gross margin percentage since we merged the companies in 2012. Our June quarter financial results came in above the midpoint of all of our guidance ranges, with earnings per share actually exceeding the guidance range. For our 2025 fiscal year, we had record revenue of $18.4 billion and gross margin of 48.8%. Our free cash flow generation in fiscal '25 was 29% of revenue and approximately $5.4 billion, which was also a record for the company in dollar terms. We're delivering on the profitability objectives discussed at our Investor Day earlier this year through a growing top line, favorable mix, and strong operational execution. First with the details of our June quarter results. Revenue came in at $5.17 billion, which was an increase of 10% from the prior quarter. The deferred revenue balance at the end of the quarter was $2.68 billion, which was an increase of approximately $670 million from the March quarter. This was related to customer advances on payments from several newer customers. From a market segment perspective, June quarter systems revenue in the foundry segment represented 52% of our systems revenue, an increase from the percentage concentration in the March quarter of 48%. In dollar terms, this level represents a second consecutive record quarter, and we also set a new record from a fiscal year perspective. We benefited from continued momentum in leading-edge processes as well as investments in mature nodes by domestic Chinese customers. Foundry was 41% of systems revenue, a slight decrease from the prior quarter level of 43%. Non-volatile memory came in at 27% of our systems revenue, which was higher than the March quarter's level of 20%. We continue to encourage you to think of NAND investments focused primarily on upgrades, which we anticipate will require an investment of roughly $40 billion over several years. DRAM decline from the March quarter coming in at 14% of systems revenue compared with 23% last quarter. The decline in the June quarter was related to the timing of certain customer projects. For the 2025 fiscal year, DRAM revenue reached a new record in dollar terms, with spending focused on technology upgrades to the 1-beta and 1-gamma nodes enabling DDR5 and LPDDR5. High Bandwidth Memory was also a key investment area. Our logic and other segment came in at 7% of systems revenue in the June quarter, slightly lower than the prior quarter level of 9%. Now I'll go through the regional composition of our total revenue. The China region came in at 35%, an increase from the prior quarter level of 31%. We saw increasing investment from global multinational customers in this region to the highest level since the December quarter of 2022. The majority of our China revenue, nonetheless, continued to come from domestic Chinese customers. The next largest geographic concentrations were Korea at 22% and Taiwan at 19% of revenue in the June quarter, both of which were a decrease from 24% concentration in the March quarter. Japan revenue at 14% was a record for Lam in dollar terms. The Customer Support Business Group revenue in the June quarter totaled approximately $1.7 billion consistent with the March quarter as well as the June quarter of a year ago. We had a third consecutive record quarter for upgrade revenue driven by NAND technology conversions. We also saw strength in our spares business, offset by a decline in Reliant Systems. Sitting here today, we do think we will see modest growth in CSBG for the calendar year. Let's look at profitability. The June quarter gross margin came in at 50.3%, close to the top end of our guided range and improving from the March quarter level of 49%. The increase is tied to a stronger mix and continued progress in our operational efficiencies from our close-to-customer manufacturing strategy. Operating expenses for the June quarter were $822 million, up from the prior quarter level of $763 million. This was a bit higher than our original estimate coming into the quarter, primarily due to increased incentive compensation tied to the company's improved profitability. R&D accounted for 69% of total operating expenses. The June quarter operating margin was 34.4% and near the high end of our guidance. This operating profit represents a record level for Lam in both dollars as well as percentage terms. Our non-GAAP tax rate for the quarter came in at 4.8%. As I indicated on the last earnings call, the rate in the June quarter reflects a tax reserve release tied to a statute of limitations expiration. Our estimate for the September '25 quarter is for the tax rate to be back in the low- to mid-teens range. Other income and expense for the June quarter was approximately $4 million in income compared with $7 million in expense in the March quarter. The improvement in OI&E was primarily the result of increased interest income tied to a higher cash balance as well as gains on our venture investment portfolio. As we've talked about in the past, you should expect to see variability in OI&E quarter-to-quarter. For capital return in the June quarter, we allocated approximately $1.3 billion to share buybacks, through a combination of open market share repurchases and an accelerated share repurchase program that ASR will continue to execute into the September quarter. We also paid $295 million in dividends. The June quarter diluted earnings per share were $1.33, exceeding the high end of our guidance range driven by that higher revenue, stronger gross margin performance, and the lower tax rate. The diluted share count was 1.28 billion shares, which was a reduction from the March quarter and was consistent with our guidance. We have $7.5 billion remaining on our Board authorized share repurchase program. Let me pivot to the balance sheet. Cash and cash equivalents totaled $6.4 billion at the end of the June quarter, an increase from $5.5 billion at the end of the March quarter. The main reason for the cash increase was cash from operating activities, including those customer advances on payments, which was partially offset by cash allocated to the share buyback, dividends, and capital expenditures. Days sales outstanding were 59 days in the June quarter, which was down from 62 days in the March quarter. June quarter inventory turns improved to 2.4x compared with 2.2x in the prior quarter. We're making progress in managing inventory levels, and we'll continue to work on this as we go forward. Our noncash expenses for the June quarter included approximately $94 million in equity compensation, $86 million in depreciation, and $12 million in amortization. Capital expenditures were $172 million, which was down from the March quarter level of $288 million. Spending in the June quarter was mainly centered on lab investments in the United States and Asia as well as manufacturing facilities in Asia, supporting our global strategy to be close to our customers' development and manufacturing locations. I would point out that offsetting this capital spending, we received more than $50 million in benefits through the advanced manufacturing investment credit as well as other CHIPS Act-related programs. We ended the June quarter with approximately 19,000 regular full-time employees, which was an increase of approximately 400 people from the prior quarter. We had headcount increases primarily within R&D to support the long-term product roadmap. In addition, we had increases within factory and field organizations for increasing manufacturing activities and a higher volume of tool installations. Let's look at our non-GAAP guidance for the September 2025 quarter. We're expecting revenue of $5.2 billion, plus or minus $300 million. We expect stronger China revenue driven by foundry spending in the September quarter. We're expecting a gross margin of 50%, plus or minus 1 percentage point. This guidance includes our current assessment of the direct impact of tariffs on our business. Operating margins of 34%, plus or minus 1 percentage point. And finally, earnings per share of $1.20, plus or minus $0.10 based on a share count of approximately 1.27 billion shares. So let me wrap up. In completing the first half of the calendar year 2025, I was pleased that we made solid progress on the objectives we shared at the beginning of the year. Sitting here today, as Tim mentioned, we now see WFE relatively balanced half-on-half. We continue to prioritize strategic investments that extend our technology leadership, operational efficiencies, and profitability, which reinforces our long-term value creation agenda.

I guess first question, your tool business is likely growing three times the growth rate of WFE here in calendar '25, and you indicated expectations for relative outperformance to continue in calendar '26. Is there a framework for thinking about the rank order of the key drivers of this outperformance that you could share?

Sure. I think it's all the things that we've talked about in the past. I mean, clearly, if we look at foundry/logic, I mentioned extensively the discussion of moly today, but we're also looking at other tools around the gate-all-around structure. It's things like selective etch and ALD. We still have backside power to come. That will be an area we believe of outperformance for Lam, given our strength in etch and deposition and the role that it plays there. We continue to see ourselves gaining against WFE the more that advanced packaging is incorporated across every type of device, whether it's foundry/logic, HPM, and even in NAND. We're starting to see on every NAND maker's roadmap, things like cell bonded to array or cell under array. And so really, as I look to the future, I mean, it basically is one in which deposition and etch intensity just continues to rise faster than WFE. Lam has an incredibly strong position already and a portfolio of products that are just doing great in the marketplace. And so I think if we continue to stay focused and execute, it will be those technology drivers that will carry us forward.

Very helpful. And then a question for you, Doug. In terms of gross margins, it sounds like there's some positive impact from China. So I'm curious if that continues into the December quarter. Additionally, is there a new normalized gross margin, excluding China, that we should be considering?

Yes, it's Tim. We are benefiting from a favorable mix, both customer, as well as a little bit of product. We do have some level of headwinds as I look forward. Tariffs are ticking up a little bit. I don't expect, as we get into the December quarter, we will continue to have quite as favorable of a level of mix. And so I'll be pretty direct about how I want everybody to think about the December gross margin. You should be kind of thinking about where consensus is today, which is about 48%. I think that's what you're going to see in December. I'm not going to get over the SKUs yet in terms of what you should be thinking about as we head into next year because I'm not exactly sure what the mix is going to be, C.J., but I'll give you a very direct guidance on December, which I just did.

Doug, regarding your comment about the decrease in gross margin, I understand that volume isn't a major factor. Given the observation that WFE appears stable from one half of the year to the next, it's clear you will gain market share on the system side in the latter half of the year. Can you share your thoughts on December revenue? Do you expect it to also decline like gross margin, or do you anticipate it will remain stable?

Yes, Tim, I mean, you should think about our revenue likely mirroring what we described a B2B, right? We told you we now think WFE is roughly flat half-on-half, flattish. And if you think through that, you know what March was. You know that both June and September are roughly the same revenue level so that you should conclude the December quarter will look largely top line-wise, like March did roughly.

Got it. Okay. So it is down. Okay. Okay. And then can you just talk about just there was a lot of puts and takes for next year. I know there was a pretty big capital expenditure cut from a big logic maker. But it sounds like you still feel like the bias to next year. I mean, you're not giving us the number. But if you had to, you would say that the bias to next year is up, is that fair?

We won't provide a specific number for 2026. However, we believe that the factors influencing WFE spending are largely favorable for Lam. It’s still early, and there are many projects underway, making it difficult to predict exact timelines. The key drivers for 2026, 2027, and 2028 include technologies like HPM, advanced packaging, gate-all-around, NAND layer scaling, and moly. We also have new products in areas like dry resist and EUV patterning that have been tested by our customers in their R&D facilities for several years and are ready for deployment. While we don’t control the timing of our customers' projects, we are very confident that when those projects launch, Lam will expand our total addressable market and gain market share. This could happen in 2026 or 2027, but our strategy remains unchanged regardless of the timing from our customers. We are positioned well for what's ahead.

Great job on the quarterly execution. Your China business was strong in the June quarter. It was up about 20% sequentially. Is the team still embedding about a $700 million negative impact from China in the second half of this year due to the restrictions that were put in place back in December? But irrespective of that, on top of all of this, you're anticipating a better overall China business this year, right? So relative to 90 days ago, what has changed within your China customer base? Do you think that this is a potential pull forward of equipment ahead of any potential tariffs, or just more focused on bringing manufacturing capabilities domestically just given the choppy geopolitical environment?

Well, Harlan, you stuck like three or four questions in there. Let me try. Yes, that $700 million number that was revenue we had identified to specific customers. Regulations haven't changed, so that's no different. I think when we think about the fact that WFE is a little bit stronger, and it's driven by a little bit more spending in China, it's just a little bit more spending from a handful of customers is how I would be thinking about it. It's nearly impossible for us to say it's a pull-in because of any specific reason. They're just spending a little bit more when we unpack it.

Okay. Perfect. And then maybe for Tim, as we track a lot of these next-generation AI, XPU, and GPU programs, like many of them are moving from 2.5D to 3D packaging. And on the flip side, you have memory customers, as you pointed out, gearing up for a strong migration to HPM 4 next year. Some of them are actually signaling increases in spending in the second half of this year, right, versus their expectations coming into the year. I think you guys did greater than $1 billion in advanced packaging and HPM last year. You came into this year targeting greater than $3 billion in advanced packaging and gate-all-around. But if you just single out advanced packaging, is that business coming in better than your expectation of entering this year? And is that what is also helping to drive maybe a slightly better second half shipment and revenue profile?

Yes, it's a little bit, Harlan. I think advanced packaging total is probably a little bit stronger than we expected. It's not wildly stronger. But HPM is strong, and there's probably a little bit of upside related to that as well as the China stuff we were talking about.

From a technology standpoint, the packaging schemes are becoming more complex. As you mentioned, the shift from 2.5D to 3D is significant. When you hear 3D, think of Lam. This involves vertical scaling and an increase in etch and deposition processes. The transition from HPM 3E to 4E is beneficial for us, both in advanced packaging and front-end equipment. As die sizes and cell sizes increase due to a higher number of TSVs required for greater I/O counts, we estimate that moving from 3D to 4D requires around 30% more wafers to produce the same number of bits. With the growing performance demands in AI, we're witnessing increased wafer fabrication equipment in etch and deposition. Additionally, when discussing storage speeds and packaging capabilities, these advancements directly enhance SSD performance and run speeds. Overall, these packaging innovations are being utilized to enhance performance and support next-generation capabilities.

I had two of them for Doug. I just want to clarify one thing on China sales. If I heard you right, you said the multinationals grew relatively more in June versus March. If true, do you think it was because of potential for restrictions on tools for MNCs in China? Or do you think something else is going on?

No, Krish, again, it's hard to isolate, hey, spending was a little stronger; what was the reason? I don't know that I would specifically identify it to that. But yes, you picked up on the commentary exactly right, the global multinationals in China grew by more than 90% quarter-over-quarter. So there was a big increase in spending from that, and that contributed to part of the uptick you saw in the China regional spending.

Got it. Got it. Very helpful. And then a follow-up for Tim. Tim, on the 2-nanometer gate-all-around, especially the leading Taiwan foundries, are all the tool decisions already made, or do you think there are still some PTORs that are still in flux?

Well, Krish, any one specific customer where they are in timing. But I would say that look, I mean, when we're not the guy in position, we're fighting right to the end until production fabs are built. But I would say that from the big drivers of SAM expansion and share gains, we've been looking forward past 2 nanometers for quite some time and a number of things I talked about in terms of inflections are beyond 2-nanometer. I just highlighted moly adoption and foundry/logic; there's a lot coming beyond, and we're already well engaged with those more advanced applications. So probably the best I can say without talking too much about one customer.

First, I wanted to zero in again on China. So the multinational is clearly the source of the upside in the quarter. Is it the local spend that's the upside in the September quarter in the multinational you see sustaining? And I guess given all of that, given the decline you're guiding for, for December quarter, is that pretty much just kind of normalizing into December? Is there something else going on there?

Stacy, I'm not going to break down the specificity of the regional composition for the guide, but China is up in the quarter. And yes, you're thinking about December in the right way.

Got it. So there's nothing else going on unusual in December. It's mostly just the China normalization? I mean the gross margin guidance seems to indicate that as well.

Yes. Revenue is going to be a little bit softer in December, just normal profile. Mix is a little bit softer in December. We're kind of back to a little bit of a run rate. And frankly, Stacy, you should also be thinking about tariffs, right? Tariffs are a little bit higher in the December quarter than they are in September. So there's a lot of moving pieces, I guess is why I'm rambling on here a little bit.

Yes, I hear you. And for my follow-up, I just wanted to ask about Taiwan. I thought you said foundry was at a record level, but Taiwan was actually down sequentially. I guess I'm just having a little bit of trouble squaring that. What am I missing?

No, Taiwan last quarter was 24% of revenue. This quarter, it was 19%. Revenue was up. So on a like-for-like basis, Taiwan was down a little bit. But understand Taiwan is not the only geographic location where there's foundries in the world, right? There's foundries all over the globe.

Where are you seeing the foundry spending picking up then?

I'm not going to break down the geographic distribution, but there's a leading-edge foundry investing in Japan. There's trailing edge foundry spending in China as well as globally. So it's a little bit all over, Stacy. Japan, I would point out to you, I mentioned this record revenue in the Japan region. So I think you probably know there's a large new foundry in Japan.

Relative to your outlook on 2026, I realize it's very early, and you don't want to give a view there. But I mean, do you have confidence that Lam's business can actually grow in 2026 even if CapEx is not up for the broader industry?

Jim, we're not going to give a number for next year. The important thing, and I think Tim described this well, listen, etch and dep as a percent of total WFE we see growing, and we feel extraordinarily good about the strength of the product portfolio right now such that we reiterated, Tim reiterated today that of this expanding SAM, we're going to gain 50% of it is our view of things. At the end of the day, we're only halfway through '25. So we're not going to quite stick our neck out about '26 yet. It will be what it will be, but we feel great about our relative outperformance into the next several years. That was the important message that we try to deliver.

Understand. Well, I had to give the rookie try. Just a second question as a follow-up. Just wondering, obviously, NAND was a pretty good step-up in the quarter. Do you see that strength sustaining through the end of the year or the next couple of quarters from where you stand today? Or is it sort of like a pop up and pop down potentially?

See, Jim, I'm also not going to get into the quarterly breakdown of NAND spending. The important thing, though, and we reiterated this, and we said it back at the Investor Day is you should think about NAND over the next several years, needing to spend roughly $40 billion to work through technology conversions, upgrades, if you will. Our view of that hasn't changed. That's the most thing to think about. The spending profile over the next several years, we see continuing.

I would like to emphasize that while we are focused on upgrades, we are also making significant advancements in technology. The demands of AI for storage, speed, and density are driving the adoption of various innovations I previously mentioned, such as moly adoption and CBA, as well as the need for QLC to increase density. I also mentioned a recent success in cryo etching, which produces a vertical high aspect ratio dielectric memory hole, an essential capability for enabling QLC. We are witnessing multiple technology drivers at play here. Furthermore, in terms of layer count, we have discussed several other factors, including backside deposition for stress management and our carbon gap fill for tier stacking. Ultimately, as we advance packaging techniques and stack cells on top of each other to achieve higher layer counts, we will see numerous upgrades to the existing systems, along with new tools being introduced to support the necessary technology advancements for improved performance and cost efficiency.

Doug, you talked about modest growth in CSBG this year. I understand you don't have to talk about WFE next year, but is it safe to assume that the CSBG business snaps back next year in a more meaningful way, given the restrictions and all that has happened in China over the years it normalizes?

You guys all led us to give you next year, and I'm just not going to do it, Atif. Listen, the way to think about CSBG though is consistent with how you should be thinking about it over the last several years, right? Chamber count grows every year. So that creates incremental opportunity for us to kind of grow spares, upgrades, service and so forth. We're really excited about the advanced servicing. Tim talked about Equipment Intelligence and cobots; that's cool stuff. We're super-gazed about that. I'm hopeful that it does better next year, but I'm just not ready to give you specificity. But the tailwinds you've always seen in CSBG continue to be there.

I want to add that the strong performance in CSBG can manifest in ways beyond just CSBG revenue. We often discuss advanced services, and I highlighted the advantages we anticipate from Equipment Intelligence and our Dextro cobots, such as improved machine availability and enhanced maintenance cycle consistency. These factors ultimately influence how customers perceive our tool as the most production-worthy system, not only for the current generation of manufacturing but for the future as well. We view the CSBG business and our systems business as highly synergistic. The success we achieve in advanced services will translate into greater market share on the systems side. Therefore, we are investing in advanced services with this synergy in mind, not solely for its revenue-generating potential.

I wanted to ask on DRAM. Obviously, AI is super strong. I think there's a lot of concerns about maybe some inventory in the HPM side. Obviously, it's a smaller business for you and lumpy, but just kind of curious what you're seeing in DRAM and any perspective for the rest of the year?

Yes, as we look toward the second half of the year, I would like Doug to add that we have noticed some strength related to HPM. HPM is indeed a significant focus in DRAM at this time. However, while it has been strong, we believe there's still a long road ahead. According to our data, approximately 7% of total DRAM bits are expected to be HPM by 2025. We can't predict exactly how this will evolve, but there seems to be a sustained demand for HPM. I mentioned earlier the effects of transitioning from HPM3 to HPM4E and potential future developments, and how these changes are influencing the number of wafers needed to produce the same bit count, ultimately increasing overall WFE. Additionally, etch processes are becoming increasingly important as we need to ensure greater precision in manufacturing these advanced DRAMs. It's also important not to underestimate the significance of our recent Akara wins in the DRAM sector. Although it hasn't been a major revenue source for us, we are capturing more market share in DRAM, and this is particularly encouraging as we expand in a growing market. Therefore, we are generally optimistic about the momentum in the DRAM space right now.

And then just wanted to ask, Doug, on gross margin. For December, is there anything more than the geographic mix? I think you said product mix, but then it would seem like China falling off a headwind as well. But maybe you can just clarify what you mentioned the product.

Yes. There's customer mix, there's a little bit of product mix, and frankly, there's a little bit of tariff showing up incrementally in December. Those are the things to be thinking about, Blayne, and frankly, overall revenue levels too. I think I told you December is going to be down a little bit. So there's a lot of moving pieces in gross margin. All of that contributes some portion to it.

Got you. So we see customers, it's related to the geographic mix.

There's some smaller customers in China, and that tends to be what I'm describing when I say customer mix. It's not specific to any one region. It's just they're smaller customers.

Just wondering on the WFE side. I think you have ITC, the investment tax rate going to 35%. Do you see that driving a tailwind to WFE as you look at next year? And then I have a follow-up.

Vijay, not that I can specifically correlate, might there be? There might be. But I've not deeply sat down and thought about this or tried to correlate; it's probably something I need to do.

Got it. And then on the China side, obviously, good to see the pickup there. As you look out, do you expect that mix to kind of normalize kind of means revert to where you guys were might be early in the year or last year?

Yes. It's a good question. I told you, I think September is up in China. I think December is probably going to lighten up a little bit. Previously, we described the view that last year to this year, China as a percent of total mix was going to be down. I think it's probably going to be flat to maybe slightly down. It got a little bit stronger, and that was part of why Tim described an uptick in WFE.

Congrats on the strong quarter. This is the third consecutive quarter where you've not only beat numbers, but the guidance has also exceeded consensus by double digits. So I guess, taking that into context, I mean, if you look at that record, I mean, is it just more conservatism in your planning? Or what has kind of surprised you by this magnitude?

I guess, Ed, if you unpack it, revenue came in a little bit better. Gross margin came in a little bit better. And frankly, that tax rate came in a little bit lower. And I don't know that I would describe a conservative bias, but that's just kind of how the quarter unfolded. Now you might point out that, hey, that happens at Lam more often than not. Yes, maybe, but that tax rate was lower than we expected for sure. It's just a little bit of all of it.

Yes, I think that while I don't want to focus too much on our conservative approach, we are in an environment where many of the markets we are targeting are showing strong demand. It's hard not to mention AI, but the truth is that demand in this area is generally surpassing expectations and driving the need for chips. Since many of the more advanced needs are related to etch and deposition, we are performing well in those areas. We are gaining a clearer understanding of how these technology transitions are happening, which is giving us confidence as we communicated during our Investor Day earlier this year about our ambitious goals for 2028 and the $1 trillion semiconductor industry.

Thank you for your perspective, Tim. My next question is about the mobile market. What are your thoughts on that sector? Carriers have recently reported very strong handset sales, with Verizon and AT&T showing a 20% increase in upgrades after a long period of stability. It's difficult to determine the reasons behind this. One carrier referred to it as pull-ins, while another disagreed. However, it seems like we might be witnessing a significant shift. If we are indeed at the beginning of a larger handset refresh cycle, how would that impact Lam in terms of logic, DRAM, and NAND? Can you quantify your exposure in this area? Also, are your average selling prices in that segment above or below your corporate average?

Yes, it's a great question. We're somewhat disconnected from smartphone and PC sales. However, I do see a slight growth in mobile and a bit of growth in PC sales. Content is really expanding. Looking at the new phones released this year, there's an increase in DRAM and NAND, and we're seeing larger base span chips as AI gains traction, which is also true for PCs. Although PC units are only up slightly this year, there is significant content growth, with some PCs now featuring 2 terabytes. This trend is evident with our large foundry customers at the leading edge and with DRAM customers producing high-performance memory, as well as NAND customers offering SSDs. That's how this growth manifests in our business.

Let me ask a few questions. First, is the Vantex you mentioned for a 400-layer application?

I'm not going to comment on which technology node it is, but when I mentioned multigenerational, you can understand that as everything from the current generation to the next few generations. It's an important win because customers tend to choose a new type of tool with the expectation that they will upgrade that tool for multiple technology nodes. So yes, it's a significant win for us.

And maybe just a quick follow-up on that. So Tim, do you think that the selection of it to be implemented at more recent node suggests maybe like a faster ramp upward in terms of vertical scaling?

Vantex has been in the market for some time now. There has been considerable discussion in recent years about leadership in the dielectric high aspect ratio etch segment, especially concerning NAND. Lam has introduced a significant amount of innovation in this area. However, our performance does not directly influence how quickly customers scale; this is dictated by their own market demands. Currently, a key area in NAND is QLC. If you have an effective cryo etch process that yields a highly vertical, high aspect ratio memory hole, the chances of successfully creating a QLC device increase. Our approach at Lam is to collaborate closely with our customers to develop technologies that empower them to succeed. When they achieve success, they tend to scale quickly. This trend is not only observable with Vantex but also across various other developments, such as enhanced read/write speeds and the rapid integration of moly in NAND. This exemplifies our role in enabling technology advancement.

Great. So maybe like SSD, very worthy. Perhaps a quick follow-up question. On CSBG, just to slightly break it down, upgrades are likely up quarter-on-quarter. This overall is up a little quarter-on-quarter. Do you think that, coming off the June quarter, Reliant has mostly stabilized or bounded out here? Even if you're not signaling for some recovery in that segment of the business?

Yes, Brian, I'm not going to get in the habit of decomposing a forward-looking statement on CSBG. It is stronger than we previously thought about in total, right? Previously, we've been describing we thought it was going to be flattish. I know we said we think it would be modestly up. So that's good. Upgrades are extremely strong, servicing spares are doing well and Reliant is down right now. That's kind of how to think about it.

Doug, you mentioned the tariff environment is very different in the December quarter versus the September quarter. I was curious if you could unpack that a little. Are you referring specifically to 232, any of the country-specific tariffs? What are you seeing as the most impactful quarter-over-quarter, September, December?

I didn't really specify the word you mentioned. There are more tariff challenges in December compared to September. That's all we conveyed. That's what I was trying to express.

Got you. And then on the customers in China, I think Stacy went through it with you guys, but more specifically, like going into September, you obviously felt like there would be a fall off. And in a very short window, those customers came back. Is December really a view of conservatism as you expect this as a one-time kind of comeback from the China up? Or in Q4, is this more of a conservative outlook? Or do you think that you could see customers stepping back in? Because obviously, the lead time here seems like it's relatively short to service these guys.

Tom, when we describe the business, we don't try to be conservative or aggressive; we call it like we see it, and that's exactly what we're doing right now.