Q1 2026 Earnings Call Transcript

Good day, and welcome to the Sandisk First Quarter Fiscal 2026 Earnings Conference Call. [Operator Instructions] Please note this event is being recorded. I would now like to turn the conference over to Ivan Donaldson, Vice President of Investor Relations. Please go ahead.

Before we begin, please note that today's discussion will contain forward-looking statements based on management's current assumptions and expectations, which are subject to various risks and uncertainties. These forward-looking statements include expectations for our technology and product portfolio, our business plans and performance, market trends and opportunities and our future financial results. We assume no obligation to update these statements. Please refer to our annual report on Form 10-K and our other filings with the SEC for more information on the risks and uncertainties that could cause actual results to differ materially from expectations.

We will also make reference to non-GAAP financial measures today. Reconciliations between the non-GAAP and comparable GAAP financial measures are included in written materials posted in the Investor Relations section of our website. With that, I'll turn the call over to David.

Thanks, Ivan. Good afternoon and thank you for joining Sandisk's First Quarter Fiscal year 2026 Earnings Call. Sandisk delivered a strong quarter with revenue of $2.3 billion, up sequentially 21% and non-GAAP earnings per share of $1.22. We generated $448 million in adjusted free cash flow and closed the quarter in a net cash position of $91 million. As discussed in February during our Analyst Day, we are focused on growing revenue, expanding margins and generating sustainable free cash flow to create shareholder value and we are executing that plan.

Our results reflect the strong execution by the Sandisk team in an environment marked by strengthening demand across our end markets. In the first quarter, demand for our NAND products continued to outpace our supply, a dynamic we expect to persist through the end of calendar year '26 and beyond. In response, we are making strategic allocation decisions to maximize long-term value creation. We are focused on advancing our technology road map and strengthening customer partnerships to deliver the right products to the right applications and customers.

Customers are proactively seeking long-term commitments given the critical nature of our technology and to secure continued access to our products. These priorities are expected to deliver durable and attractive financial results while unlocking the strength of our broad product portfolio. With investments in data centers and AI infrastructure expected to surpass $1 trillion by 2030, the demand for NAND storage products capable of processing large volumes of data quickly and efficiently is increasing dramatically, creating a strong tailwind for our high-capacity power-efficient SSDs enabled by our BiCS8 technology.

BiCS8, which delivers industry-leading capacity, I/O performance and energy efficiency, accounted for 15% of total bits shipped and is expected to reach majority of bit production exiting fiscal year '26. Our BiCS8 products are expected to enable us to grow our data center business while further strengthening our positioning in the edge and consumer markets. Let's dive into the first quarter results by business. Our data center business gained momentum with revenue up 26% sequentially as global hyperscaler, neocloud and OEM customers are seeking to deepen their partnership with Sandisk. Our storage-focused SSD product line, code named Stargate is growing in demand with 2 hyperscaler qualifications underway and a third hyperscaler along with a major storage OEM planned for calendar year '26. Across the data center portfolio, we are working with 5 major hyperscale customers through active sales and strategic engagements.

In edge, we are seeing positive momentum from a PC refresh cycle, aided by Windows 11 adoption and Windows 10 end of life. PC unit shipments are expected to grow low single digits with mid-single-digit growth in capacity per device in calendar years '25 and '26. Beyond PC, premium smartphones are delivering modest unit growth, supported by new model launches featuring enhanced generative AI capabilities. Average smartphone capacity per device is expected to grow high single digits in calendar years '25 and '26. Looking ahead, we expect continued momentum in edge as device upgrades accelerate, driving increasing NAND content.

Ongoing supply and demand dynamics are expected to extend the need to strategically place bits as mentioned above. As customers across data center and edge seek higher performance AI inference capabilities, demand for innovative solutions to address AI inference storage has increased interest in our high-bandwidth flash or HBF technology will deliver. Building on the technical advisory board and ecosystem partnership with SK hynix we announced last quarter, we are actively engaging potential customers for inference applications in both data center and edge. As we enter the holiday period, we are also well positioned to capture strong seasonal demand with our refreshed consumer portfolio and significant presence across key retail and online channels.

We are engaging with the gaming and creator communities to sustain our momentum. Our recently launched Memory Man campaign is creating lots of interest and excitement, strengthening brand relevance ahead of the holiday season. Also in consumer, our partnerships with leading companies like Nintendo remains strong with solid adoption of our co-branded Switch 2 microSD Express Card, which eclipsed 900,000 units sold in fiscal Q1. We are also expanding our presence in the handheld gaming sector with the new Sandisk microSD for ROG Xbox Ally, reinforcing our position in gaming storage. Our consumer business remains a major focus for the company, driving revenue growth and attractive margin through cycle.

In summary, this is a new era for Sandisk. We have a strong balance sheet, an industry-leading product portfolio and a clear technology road map that drives organic strategic customer engagements. As we help power one of the most transformative technology megatrends of our time driven by accelerated AI proliferation, we are confident in our ability to create significant and sustainable value for our customers and shareholders. With that, I'll turn the call over to Luis to dive deeper into our financial performance and guidance.

Thank you, David. Let's dive deeper into the quarter results. Revenue for the first quarter was $2,308 million, up 21% quarter-over-quarter and up 23% year-over-year. This compares favorably to our guidance of $2.1 billion to $2.2 billion.

Bits were up mid-teens sequentially with pricing up mid-single digits. Pricing strengthened during the quarter. Higher-than-expected bit growth enabled the revenue over delivery. Before reviewing the details by market, I will share that we will be aligning the names for our end markets to match the nomenclature commonly used in the industry.

Going forward, we will use data center to refer to the business that's comprised primarily of our products for public and private cloud environments. We used to refer to this business as cloud. We'll use edge to refer to the business that serves our original equipment manufacturer and channel customers with a broad array of high-performance flash solutions across computer, mobile, gaming, automotive, virtual reality headsets and other edge devices. We used to refer to this business as clients.

We will continue to use consumer to refer to the business that contains our broad range of retail and other end user products, which capitalizes on the strength of our product brand recognition and vast point of presence around the world. In the first quarter, we saw strong sequential demand across all end markets. Edge revenue came in at $1,387 million, up 26% sequentially. Consumer revenue came in at $652 million, up 11% quarter-over-quarter and data center came in at $269 million, up 26% sequentially.

Non-GAAP gross margin for the first quarter was 29.9%, up 350 basis points quarter-over-quarter. This compares favorably to our guidance of 28.5% to 29.5%. The incremental revenue drove the higher-than-expected gross margins. In the first quarter, we incurred $61 million in start-up costs and $11 million in underutilization charges.

Excluding this cost, non-GAAP gross margin would have been 33.1%. Non-GAAP operating expenses for the first quarter were $446 million, which is higher than our guidance of $415 million to $430 million. Operating expenses were above guidance, mostly driven by higher variable compensation from the revenue over delivery versus plan. As a result, non-GAAP operating margins at 10.6% were up 530 basis points quarter-over-quarter.

Non-GAAP EPS for the first quarter were $1.22, up from $0.29 in the prior quarter. This compares favorably to our guidance of $0.70 to $0.90. The non-GAAP EPS beat reflects a higher-than-expected revenue and gross margins and a more favorable tax rate. Key GAAP to non-GAAP reconciliation items include $47 million in stock-based compensation, net of taxes, which represents 2% of revenue, $9 million in separation charges and $17 million in onetime costs related to the SSDs transaction and separation from Western Digital. Moving on to the balance sheet.

We closed the quarter with $1,442 million in cash and cash equivalents and $1,351 million in gross debt. We achieved a net cash position approximately 6 months faster than the target shared during Investor Day in February, driven by strong cash focus in a robust market. During the quarter, we paid an additional $500 million of our TLB and reduced our inventory days from 135 to 115 as demand exceeded supply. Moving on to free cash flow.

During the quarter, we generated $448 million in adjusted free cash flow, which represents 19.4% free cash flow margin. This included $488 million cash from operations and $10 million cash received from our activities related to Flash Ventures, partially offset by $50 million invested in our back-end operation and offices. The $10 million received from our operations related to Flash Ventures includes $337 million in gross CapEx with $107 million funded through depreciation as part of our cost of goods sold and $240 million funded from external sources, mainly subsidies and equipment leasing. Altogether, our gross capital expenditures totaled $387 million and represents 16.8% of revenue.

Moving on to guidance. For the second quarter, we expect revenue between $2,550 million and $2,650 million due to double-digit price increases and mid-single-digit bit growth. Consistent with our expectations, we anticipate demand for our products to exceed supply throughout the end of the calendar 2026. Based on current supply and demand dynamics, we believe demand for our products will exceed supply beyond that period. Our products are currently on allocation across all end markets.

Recall, the third quarter is a seasonally lower volume period for our consumer business following the holidays. Our forecast for non-GAAP gross margin for the second quarter is between 41% and 43% from higher pricing and cost tailwinds. This estimate includes an expected $30 million in start-up costs. For the second quarter, we expect non-GAAP operating expenses between $450 million and $475 million.

The incremental operating expenses are to support our data center business expansion and our HBF innovation. We expect non-GAAP interest and other expense between $40 million and $45 million and non-GAAP tax expenses between $80 million and $90 million. We forecast non-GAAP EPS for the second quarter between $3 and $3.40, assuming 155 million fully diluted shares. The higher diluted share count in the second quarter is driven by the increase in the stock price following the treasury model.

We expect to generate positive free cash flow in the second quarter despite capital investments to enable the BiCS8 transition, which is expected to be our most significant node by the end of the fiscal year. Our fiscal 2026 CapEx plans remain unchanged along with the long-term strategy to grow supply in line with the market, assuming bit demand compound annual growth rate in the mid- to high teens. Our capital allocation priorities are consistent with what we shared during Investor Day in February. Our first priority was to achieve a net cash position, which we have now accomplished through strong cash generation.

Going forward, our capital allocation is unchanged, and we expect to continue to invest in the business and return cash to shareholders. We're executing the strategies and plans that we shared with you in February, and the results are coming in as anticipated with revenue growth, margin expansion and more efficient use of assets. We remain focused on creating sustainable value for customers and shareholders with continued prudent management of the business. With that, let me turn the call back to David.

Thank you, Luis. In summary, Sandisk delivered a strong start to the fiscal year, underscoring the success of our strategy to drive profitable growth, expand margins and generate sustainable free cash flow. Our solid execution amidst robust demand positions us well for continued momentum across the data center, edge and consumer markets. Our ongoing qualifications and strategic engagements with key hyperscale customers underscore the growing adoption of our high-capacity power-efficient enterprise SSDs.

As we move through the remainder of fiscal 2026, our focus remains on disciplined capital allocation, operational excellence and delivering differentiated technology that meets evolving customer needs. We are well positioned with industry-leading technology and products, healthy long-term market fundamentals, a strong balance sheet and an efficient operating model to create substantial value for our shareholders. Our technology is arriving at exactly the right time. When the market is ready, the demand is real and the opportunity to drive meaningful earnings power is just starting.

With that, we will open the call for questions.

[Operator Instructions] Our first question comes from C.J. Muse with Cantor Fitzgerald.

I guess, Dave, you spoke to customer engagement evolving now that we're on allocation. And you know very well how that's transpired in the HDD world moving from build to order, followed by long-term agreements. So curious, are you seeing similar trends emerge here in NAND? How does your visibility extend? And how are you allocating the bits that you can produce?

Yes. So I'll let Luis talk about the allocation. But yes, -- C.J., it's good to hear from you, first of all. Thanks for being on the call.

I would say there's kind of 2 phases of how we're hearing from customers and what they're reaching out about. There's a phase where we're striking deals that are multi-quarters, let's say, through the first half of next calendar year that are volume and price kind of deals where customers are looking for certainty of supply. And so that's a little bit different than what we've seen in the past where everything is usually just quarter-by-quarter based. But especially as data center starts to grow, those customers are reaching out proactively and providing visibility all the way through calendar year '27 of what their demand is going to be and kind of want to have conversations on how we could line up our supply to that demand given the product -- the emerging product line we have that's under qualification across all those players.

So it is a very different time. I'd say it's still fairly nascent, especially that second phase of it, and we'll be having those conversations over the next several months and keep you updated on how that goes. But I think it's certainly a very welcome development in this market. As you know, this is a market where we have to make very long-term CapEx decisions.

And demand over the next quarter or 2 or 3 doesn't really move the CapEx number. We're looking for what demand is going to be over the next many years. So having these conversations with what is emerging as the largest customers in the market is, as I said, a very welcome development. We'll be engaging in those conversations more deeply over the next several months.

So Luis, do you want to talk about allocation?

Yes. C.J., we're working very closely with our customers. We talk to them all the time. And what we're doing is we're prioritizing our most strategic customers, those customers that have been very close to us, those customers where we see growth, those customers where we can create value for them and value for us, right?

And that's how we're evolving our portfolio. We're going from a mobile-centric company to really serve our customers, and we're seeing very strong growth in our data center business. So we're very excited about the opportunity that this presents to us, but working very closely with our customers, particularly those more strategic to us.

As a quick follow-up, how are you thinking about your bit shipment growth opportunities here in calendar '25, '26? Your days inventory down to only 107 days and BiCS8 is just kind of ramping. How are you thinking about what kind of bits you can ship this year and next?

Yes. Our goal is to keep our market share, right? We're not here to disrupt the market, but we're very optimistic about where things are heading. We're doing well in our client portfolio.

We're doing very well on the consumer business. And we are very optimistic that we can build our business in data center where we are underrepresented, and we've talked this in previous meetings. So -- and the way to do it is through innovation. We have BiCS8 coming in.

We're getting qualification with our products. So we feel good about our market share on bits across the market.

So C.J., well as Luis -- just to add a little bit. I mean, Luis, as he rightly said, we're looking, and we plan to grow along with the market. And we've got the capital plans to support that. We're investing assuming that mid- to high teens level of demand on the long term.

And then as demand exceeds that, we'll talk about the allocation phases that we talked about. But as I said earlier in the first part of your question, we're optimistic that our customers are starting to look further down the road as far as what their demand is. And I think that helps this whole equation. But we're definitely growing along with the market.

We have our capital plans, and they remain unchanged.

Our next question comes from Jim Schneider with Goldman Sachs.

Just structurally, obviously, we're in a supply-constrained scenario, and I think a lot of customers are clearly asking for supply as you kind of talked about. So maybe you just kind of give us your view on over the next couple of years, the supply situation you expect to deliver, whether that's just through upgrades at this point? And what would make you decide to add wafer capacity over the next coming years?

Yes. So first of all, on the overall environment, I think we've been talking for, I don't know, at least a year now that we saw an undersupplied market through the end of '26, and that's mainly focused on looking at long-term demand trends and then what capital has been invested in this business over the last several years and factoring in nodal transitions of all the players in the market. So as we said in the prepared remarks, we're now seeing that push out beyond '26, just given these conversations we had earlier where customers are coming to us looking for '27 supply. So we've kind of planned for this market.

It's hard to call sometimes on a quarter-by-quarter basis, but we've seen this set up for quite some time. As far as what does it take, we're not at the phase of talking about additional capital in this business. We're investing a significant amount of capital, as Luis said, to do the BiCS8 transition. We have a very, very strong technology road map where we can increase productivity and bit supply without increasing wafers.

When you start to get into that kind of discussion, which we are not yet quite frankly, it's where we -- we need to -- it's not about what demand is going to be next year or the year after that if we're going to add capacity, we got to see demand for a long period of time. So we have zeroed in on that mid to high teens, clearly, we are above that right now, but that's a long term number we are investing to.

And then maybe relative to your position in enterprise SSDs. I think you gave some targets at the time of separation in terms of your ambitions in that market. Maybe give us an update on how the qualifications are going and as you look into the end of 2026 or 2027, what percentage of market share you might hope to attain and what -- how big of a part of the business that might be?

Yes. So we're very, very happy with where the business is. It's driven by where the product portfolio is. We talked about our Stargate program, which is storage-based enterprise SSDs.

We talked about that at our Analyst Day. Those products are now in customers' hands and starting qualifications, the 128T drives and we'll be moving up from there. Those are the part of the products where people are talking up -- talking to us already about supply several years out. So they feel good about the products.

We feel good about the products. We also have our compute-focused enterprise SSD, which continues to do very well, and the number of customers deploying that continues to broaden. So all of the right things are happening to move the portfolio in the right direction and move how much we're shipping in the right direction. Look, I think you're going to see -- our plan is you're going to see increasing sales in this segment sequentially throughout FY '26. I think we'll end the fiscal year in a strong position from an exit rate point of view.

It's a bit of -- it's a difficult market right now to completely start talking about market share because the market is very dynamic. The market is -- it's almost like every week or 2, our estimates for calendar year '26 demand in the data center market move around and they're all moving up. When we were sitting here 3 months ago, we thought our forecast was data center exabytes would increase mid-20% level in '26. We've now upped that to mid-40% in '26. So the market is moving very quickly. And what our goal is to get our fair share of that market.

We think the product portfolio is exactly in the right spot. We think BiCS8 is absolutely the right node to drive that across performance and density. And you're going to see this whole story play out in the numbers as we move through calendar year '26.

Our next question comes from Aaron Rakers with Wells Fargo.

I guess my first question is just building on the last one, David, there's been a lot of discussion around hard disk drives being supply constrained and what seems to be kind of an inflecting AI narrative around enterprise SSDs. I'm curious how you assess kind of the enterprise SSD market opportunity relative to hard disk drives. Is that -- has your thought process changed at all? Have the shortages in hard drives on a nearline perspective driven any kind of change of engagements on those opportunities?

Yes. I mean my primary view on this, Aaron, as you know, we've talked about this for a long time, maybe years and years going back to when we ran both franchises is these are primarily complementary technologies in the cloud. And we're definitely seeing kind of the rising tide lifting all boats. And I do think you have a dynamic that with AI, more data is getting warmer and warmer data moves to enterprise SSDs.

So I've always thought that both technologies are going to grow, and enterprise SSD is going to grow faster. Is it a substitution question? Is it one or the other based on some shortages or tightness in one market versus the other over time. There could be some of that dynamic going on.

But I think the long-term drivers for enterprise SSD are extremely strong in the data center. We're seeing more data being stored all the time. The value of data is going higher as it's being used to train more and more models. And now we're seeing the video creation models getting much, much more sophisticated and just driving even faster creation of data.

So obviously, we're very bullish long term on the data center market. I think the NAND market is going through a very interesting transition right now. I mean calendar year '26 will be the first time that data center market is the largest market in NAND. That's always been the mobile market.

And so we're seeing a major inflection there. The growth rate is higher. It's now going to be the biggest segment of the market. And it's also -- from a customer point of view, it's a more diverse market.

So I think that changes the dynamics of the way the conversations with customers, the way purchasing decisions are made, pricing, visibility. All those things, I think, are changing before our eyes as the data center market emerges as the largest market in NAND. And the corollary to that is the customers behind that are -- if you look out at, there are 27 exabyte demand numbers we're talking about, they're very, very big numbers. So I think all that says, it's a very good market, very -- better visibility, growing visibility for enterprise SSD.

Is some of that because of what's happening in HDD? Maybe it is, but I don't think that's the primary driver of what's going on here and what's going to happen over the next several years.

Yes, that's very helpful. And then Luis, if I can, real quick. I mean, in your prepared comments, you alluded to seasonality or just be aware of seasonality into the March quarter. Can you kind of unpack that?

What should we be thinking about? Are we down sequentially in the March quarter? It seems like pricing could continue to trend higher. And how do we think about kind of feathering out the start-up costs as we move forward?

Yes. So let me start with the last part of your question. Start-up costs, consistent with what we said in the last call, we went from $60 million last quarter to call it, $30 million this quarter to pretty much 0 going forward. So that's easy to put into your forecast.

Now I don't want to guide into Q3, but I did want you to be careful as you build your models in Q3. If you look at historically, right? Our bits are down in Q3 somewhere around 12% to 14% sequentially. Now the dynamics in the market may be a little bit different because data center is stronger, and therefore, the mix of our portfolio is slightly different. But I want you to be careful and just assume that there is some seasonality, which is particularly important for us given that we have a stronger consumer business.

Our next question comes from Joe Moore with Morgan Stanley.

I wonder if you could characterize some of this data center demand that you're seeing. Is there demand for QLC on the enterprise side? How much of it is QLC? And can you talk a little bit about how BiCS8 enables you to kind of increase your presence in that market?

Yes. Joe, I would say -- I don't have a split for you across QLC and TLC, but there's these 2 primary use cases, which is what we -- the compute enterprise SSD, which is a TLC faster interface. We have good demand for that across the -- across our customer base. We have a lot of customers coming in looking for upsides on that.

And then you've got the storage class product, 128T is what we're qualifying, which is a QLC product. And BiCS8 QLC, very energy-efficient, high performance. So we think that node is extremely well positioned. In FY '26, I actually do have some numbers here, QLC going from 20% to 40% of the market by the end of our business by the end of FY '26. So definitely seeing strong growth in that storage category as that product gets qualified in the market.

Great. And then as you describe the market, I would have to think that the fabs are all running full in NAND at this point, but you didn't mention underutilization expense. Do you think there's any incremental supply coming from the fact that people had underutilized the fabs earlier in the year? And just how does that affect your view of '26?

Yes. We've moved to 100% utilization, Joe. So fabs are running at full capacity, and we keep on pushing them to produce as much as we can because as you've seen, inventories are down very meaningfully. So yes, we're running full capacity, Joe.

So Joe, just to put some numbers around it, I mean, we see supply growth -- we saw supply growth in calendar year '25 of about 8%. We see it at about 17% in '26. We see demand -- constrained demand around 14% because that's what -- mid-teens because that's what -- that's all that's out there from a supply point of view. But unconstrained demand is in the -- literally, a couple of weeks ago, we thought it was 20%, it's probably mid-20s by now. So we see the supply pretty much being able to service that kind of mid-teens level demand for '26.

Our next question comes from Mark Newman with Bernstein.

Congrats on a great quarter. So just digging a bit deeper into the supply-demand dynamic. It seems like things are going great. I wondered if you could give us a little bit more clarity on the portion of your contracts that are shorter term versus longer term?

Because obviously, longer-term contracts, as was discussed a little bit earlier, has a lot of positives in terms of giving you more confidence in the demand for longer term. But on the other hand, when you're talking about pricing, some of the pricing data out there is inflecting up significantly right now, but longer-term contracts may not actually inflect obviously, because they're longer-term contracts. So I just wondered if you could break out for us what portion would be shorter term, say, a quarter or less or what portion would be 6 quarters would be useful? And I have a follow-up on HBF as well.

Yes. Thank you for the question. Yes. So we don't have volume -- we have very little volume price commitments that are beyond a quarter.

So that's -- what we referred to in our prepared remarks is what we are hearing now from some of our very strategic customers, very large customers is they want assurance on supply. So they have approached us willing to start some of those conversations to see if there is a volume price commitment that we can agree for the year, potentially for longer. So we're going through that process. But today, as things stand, we have very, very little volume and price commitments that expand a quarter.

And then if you could give any more clarity on the road map for HBF, high-bandwidth flash. You discussed earlier in the prepared remarks about starting to work with some customers. Any updates on potential time line there?

So we're on -- we announced a time line last quarter of having the memory later in '26 and then having the controller for that in '27 we're still working towards that time line. We have a robust set of customer conversations going, looking at use cases, both in edge and cloud of how that product can be integrated into the architectures that customers are building across those -- across devices and the cloud. So we're very optimistic about the technology and where it's headed and the use cases, and we continue to do the work on the technology and have a good set of conversations with a number of customers about how it can be deployed and zeroing in on the use cases, which helps you really dial in all the requirements.

So work in progress.

Our next question comes from Karl Ackerman with BNP Paribas.

I have 2 as well, please. As you seek to qualify more hyperscalers on your enterprise SSD portfolio, would you anticipate simply transitioning client and edge wafer capacity to enterprise? Or would that come primarily from new capacity coming online on your K2 fab?

No. Look, we're always -- again, we're growing bits all the time because we're growing with the market, and we do not plan to add capacity for any particular market. It will simply be a mix question either on a quarter-by-quarter basis or if there's -- obviously, if we get to the point where there's longer-term commitments, then we'll have more visibility into what that mix is going to look like. But as I've talked about quite a bit, our whole goal is as much optionality as possible and then mix for the best financial return in any given quarter with the supply we have available.

Got it. For my follow-up, I guess, how should we think about cost declines going into the December quarter? And if we zoom out, is it fair to assume cost declines can approach, I suppose, high teens as you transition aggressively toward BiCS8?

Yes. We stopped talking about cost declines a while back, as you know, but we'll give you some -- a little bit of maybe runway lights around it. So we're coming out of a period where we've had quite a bit of cost headwinds in the business, and we've been pretty clear about that over the last several quarters. And now we're transitioning to the BiCS8 ramp.

We're getting the underutilization costs behind us. We're getting the fab start-up costs behind us. So in the December quarter, those cost headwinds turn into cost tailwinds, which is a great place to be. As we ramp BiCS8, we'll continue to get some cost out of it.

But again, the costs are more idiosyncratic on a quarter-by-quarter basis. And we don't want to put an overall number out there, but the numbers you're talking about are pretty aggressive, too aggressive.

Yes, Karl, our focus is obviously on driving our gross margin, and we feel good about the progress we're making here, right? We added whatever, 350 basis points quarter-over-quarter, 720 basis points in 2 quarters. Still, you would say we're 4 quarters below our model. So we need to get several quarters ahead of that number. But we're very focused on gross margin and driving that up.

Our next question comes from Mehdi Hosseini from SIG.

Two follow-ups. What's the update on the UltraQLC 256TB? I think last earnings call; you said that you have a Tier 1 data center customer that should be ramping in the first half of calendar year. And what's the update there? And I have a follow-up.

No, that was the 128 that we're going to be ramping in calendar year '26, and we'll move -- that 256 is a -- will start hitting the market mid- to late next year and ramp the following year is the most likely timing for that. Again, I want to be careful getting into product timing that's not announced. But the qualification that's going on across Stargate right now is the 128T drives.

Okay. So you have one customer for 128 that should be ramped in the first half of calendar year.

We have multiple customers under qualification now. We have customers lined up for qualifications next year and you'll see ramping as -- qualifications can take quite a bit of time, as you know, several quarters. So you'll start to see the ramp of that product mid next year. Across all of our enterprise SSD portfolio, we're working with, as we said in the prepared remarks, 5 hyperscalers across all the different technology we're building.

So we feel very good about there's much broader adoption that goes -- it happens quarter-over-quarter.

And as a follow-up, assuming that there won't be much of a wafer capacity add at the JV, should we assume that as this 5 as the 128 terabyte ramps middle of next year, you would allocate more bits towards the customers, and therefore, there will be a more accelerated mix shift towards the cloud mix?

Yes. I mean we don't want to get too far ahead of ourselves on forecasting quarters. But that's -- what I said is you're going to see sequential growth in our data center portfolio throughout fiscal year '26. And so you can assume you're going to see an exabyte shift in that direction as well.

And remember, even if we don't produce more wafers, those BiCS8 wafers contain a lot more bits, right? So that productivity allows you to produce more bits and therefore, to supply more of our customers.

Our next question comes from Wamsi Mohan from Bank of America.

It's Ruplu filling in for Wamsi today. I have 2, one for Dave on bit growth. Dave, you had strong mid-teens bit growth in the first quarter. Can you talk about how you're thinking about bit growth across the different markets, data center, client, consumer in the second half?

And you talked about some products in allocation. Should we assume that's across the product line? And I'm assuming all of these orders are noncancelable. So any comments on the expected growth by end market in the second half and which products are in allocation?

And I have a follow-up for Luis on margins.

Do you want to take that one? You want me to...

I can. Yes. So we feel good about our growth across segments. Obviously, data center is the segment, the end market that's growing the fastest.

And therefore, we expect to grow there faster in the back half, as David alluded to at the beginning. But we do expect to continue to make progress on both the consumer and the edge market. So I would say our growth will be a little bit heavier on or higher on the data center end market.

Okay. Okay. That's helpful. And Luis, if I can follow up on margins.

It looks like you said in the first quarter, ex the start-up cost and underutilization, the gross margin would have been 33.1%. You're guiding at the midpoint of about 900 bps -- sorry, yes, a very strong 900 bps of growth in sequentially. So can you talk about what are the factors that are going into there? Should we expect another round of price increases?

What is the factor -- how much is mix a factor to that? And I think you said there are no more underutilization costs and you're not talking about cost down. So are these the only 2 factors? Or is there FX or some other thing also that is impacting margins, which you've got a very strong guide.

Yes. If you are comparing the 33% versus the 41%, 43%, you are comparing that already excluding start-up costs underutilization. So you're looking at it the right way. I would say the majority of that is the pricing that we're implementing.

In my prepared remarks, I mentioned that during the quarter, we saw an improvement in pricing during the quarter. So the margins at the end of the quarter were better than the margins at the beginning of the quarter, and we expect that to continue. So pricing is a key driver of that. And obviously, as we shift more into BiCS8, we'll see lower cost per gigabyte.

Again, that's not an area we want to spend a ton of time talking about, but we do see a little bit of a benefit there. So combine the 2, we feel good about our gross margin expansion, which should continue.

Our next question comes from Asiya Merchant from Citigroup.

Can I just clarify on the guide at the midpoint, I think it's 2.6%, and I think that's up like low double digits. I thought pricing in itself was up low double digits. So if you could just clarify that for us and you're still looking for bit growth unless I'm missing something here in my calculation. Then I have a follow-up.

Yes, it depends on whether you're looking at the midpoint, the high point, the low point. But yes, we expect pricing to be double digit, and we expect some low single-digit bit growth, as we mentioned in our guide. Inventory levels are low, as you've seen. That's kind of the construct, right?

So depending on which range you're looking at, that would be the combination of bits versus pricing. So I think the key message is most of the growth in revenue will be pricing driven in the quarter.

Okay. And then just for my follow-up, a lot of questions on the data center. But just wanted to -- if I heard David correctly on the edge, the client PC and the mobile phones, just the confidence as we look into calendar '26 on the growth there. I think some of the expectations on unit growth were a little bit higher than maybe some industry is forecasting.

So just if you could click on what's driving the confidence there on the edge devices, particularly PCs and mobile phones.

Yes. I mean -- so phones, we've got units up slightly, but content per unit up double digits. So that's a strong number across [ 1.2 x ] billion units. And then you got PCs basically flat to slightly up, and then we got average capacity up, mid-single digits.

So you've got -- that's good exabyte growth across those markets. And then on top of that, you've got what we've talked a lot about with the exabyte growth in data center. But there's one of the great things about the NAND market. It's very diversified.

NAND plays everywhere. There are these 3 very large pillars in the market. And we're seeing growth out of all of them and really very, very strong growth out of the data center, as we've talked about. And again, as I said earlier, it's just very interesting to see the data center now emerge as on an exabyte basis in '26 will be the largest market in NAND.

And I think that's bringing some fundamental change to the way the market works. And I think we're going to see that play out over the next couple of quarters.

Our next question comes from Steven Fox with Fox Advisors.

I just had one question for Dave. I guess I'm trying to figure out like how some of these long-term agreements might play out. I mean it seems like we're at a historic impasse with NAND and the demand coming from the cloud guys. So I don't know if you want to give any of this away, David, but like is there any kind of values that we should think about in these negotiations, whether it's sharing of cyclical risk, sharing of cash, CapEx, things like that?

Anything you could sort of talk about? And then how does that trickle down to like customers that aren't lucky enough to be able to have longer-term supply agreements?

So Steven, I think it's a little early to get into that level of detail. I think the key issue is that it's the customers reaching out to us, right? And as I've said a couple of times on this call, as data center emerges as the largest market -- and again, all the markets are very, very important and all the customers are very important. But as these really big players are looking at their road maps and looking at their businesses, they're proactively reaching out now all the way into '27 and want to talk to us about supply across our portfolio.

So it is just a different dynamic. I think we're working to have conversations with them to understand what those relationships could look like. It's very welcome. Again, as I said earlier, we have to make very long-range decisions when we spend money, and we're certainly have an enormous R&D capacity that's working on 2 or 3 nodes in the future for -- to drive growth in this technology.

So we're making very long-term investments and to see our customers proactively -- and what is becoming the largest customers proactively reaching out and talking about multiyear supply dynamics is, I think, a very healthy sign. Exactly how that's going to get worked out, I think it's a stay tuned kind of message. We're kind of in the early stages of those conversations.

Our next question comes from Krish Sankar with TD Cowen.

I had 2 of them. First one, Dave or Luis, your data center revenues grew very nicely. I'm just wondering, is there a way to figure out how much of that is related to AI data center and neocloud versus traditional cloud? Is there a way to put a percentage around it? Or is it all majority AI? And then I had a follow-up.

Yes. I would say it's majority AI-driven growth.

Got it. And then I think, David, you kind of talked about a 40% growth next year for data center exabytes and then even better growth in '27. I'm just curious, is that the right way to think about it, data centers is a 300-plus exabyte market this year going to mid-400 to 600 plus in '27?

Yes. We've got it as a high 300s number for '26. So that's healthy growth over what we saw in '25. Remember, going back to '24, we saw 130% year-over-year growth in that market. That was a 200 -- like around a 230 number, followed up by high teens in '25, and we're going to put on top of that a mid-40s number in '26. So yes, it's healthy growth in that market. And as I said, we're seeing those customers really start to change the dynamics of kind of structurally how we think about this whole supply-demand [Audio Gap].

More to come.

Our next question comes from Nam Kim with Arete Research.

I also have 2 questions on HBF. I know it's still early, but how do you see the long-term growth potential? And can you provide any qualitative color or early market sizing? And then second, how are you approaching this given competitors' greater TSV stacking experience from HBM? What competitive advantage do you think Sandisk can bring to HBF?

Yes. So I think on both of those questions, I'm going to defer. We'll talk more about the market potential and kind of the TAM numbers and all the things you're talking about. I mean, just to reinforce, this is an inference-based solution.

We're not going after the model training side. We're not trying to replace HBM in that market. But we do believe there's a very large opportunity in the inference market, especially on the device market, given the footprint of what the density we can get out of flash. As far as what is the -- what is our competitive advantage, again, I think we'll wait on that as well.

We've done a lot of work on the actual NAND design itself. And I think I said this a couple of quarters ago. I mean one way to think about it is NAND designers are always thinking about how they expand density. That's like been the #1 thing.

How do I get more bits out of how I design NAND. And when you actually start asking folks to start thinking about, well, how do I get higher bandwidth, all these kinds of questions, it turns out you can do -- how do I make the technology more durable. We've been thinking about those questions now for several years and have some design innovations, let's say, that we think make HBF a very compelling technology, but we're not ready to talk about what those are publicly.

Our next question comes from Mark Miller with The Benchmark Co.

Congrats on a great quarter and great outlook. Just with the strong sequential growth you've seen in data center, do you think you're picking up share there?

Yes. We think we're growing faster than the market is growing, right? And we think we're going to see that throughout the fiscal year.

This concludes our question-and-answer session. I would like to turn the conference back over to David for any closing remarks.

Okay. I just want to thank everybody for joining the call, all the great questions. We look forward to talking to you throughout the quarter. Thanks again.

The conference has now concluded. Thank you for attending today's presentation. You may now disconnect.