管理層發言
Greetings, and welcome to the MaxLinear Second Quarter 2026 Earnings Conference Call. As a reminder, this conference is being recorded. It is now my pleasure to introduce Leslie Green, Investor Relations. Please go ahead.
Thank you, Paul. Good afternoon, everyone, and thank you for joining us on today's conference call to discuss MaxLinear's Second Quarter 2026 Financial Results. Today's call is being hosted by Dr. Kishore Seendripu, CEO; and Steve Litchfield, Chief Financial Officer and Chief Corporate Strategy Officer. After our prepared comments, we will take questions. Our comments today include forward-looking statements within the meaning of applicable securities laws, including statements relating to our guidance for the third quarter of 2026, including revenue, GAAP and non-GAAP gross margin, GAAP and non-GAAP operating expenses, GAAP and non-GAAP interest and other expense, GAAP and non-GAAP income taxes and GAAP and non-GAAP diluted share count. In addition, we will make forward-looking statements relating to trends, opportunities, execution of our business plan and potential growth and uncertainties in various product and geographic markets, including, without limitation, statements concerning the future financial and operating results, opportunities for revenue and market share across target segments, new products, including the timing of production and launches of such products, demand for and adoption of certain technologies and our total addressable market. These forward-looking statements involve risks and uncertainties, including risks outlined in the Risk Factors section of our recent SEC filings, including our most recent annual report on Form 10-K and our Form 10-Q for the quarter ended June 30, 2026, which we filed today. Any forward-looking statements are made as of today, and MaxLinear has no obligation to update or revise any forward-looking statements. The second quarter of 2026 earnings release is available in the Investor Relations section of our website at maxlinear.com. In addition, we report certain historical financial metrics, including, but not limited to, gross margin, income or loss from operations, operating expenses, interest and other expense and income tax on both GAAP and non-GAAP basis. We encourage investors to review the detailed reconciliation of our GAAP and non-GAAP presentations in the press release available on our website. We do not provide a reconciliation of non-GAAP guidance for future periods because of the inherent uncertainty associated with our ability to project certain future changes, including stock-based compensation and its related tax effects as well as potential impairments. Non-GAAP financial measures discussed today are not meant to be considered in isolation or as a substitute for comparable GAAP financial measures. We are providing this information because management believes it is useful to investors as it reflects how management measures our business. Lastly, this call is also being webcast, and the replay will be available on our website for two weeks. And now let me turn the call over to Dr. Kishore Seendripu, CEO of MaxLinear. Kishore?
Thank you, Leslie, and good afternoon, everyone. Our Q2 financial results highlight the exciting inflection in our business trajectory and the beginning of a multiyear growth phase for MaxLinear. Overall, revenue grew 55% year-on-year, reflecting strong execution, accelerating adoption of our newest data center products, improving visibility and gathering momentum in our infrastructure portfolio. With the improvement in profitability in the quarter, we also returned to positive GAAP EPS of $0.02. In addition, with a favorable product mix shift towards our infrastructure business, we are now forecasting Q3 2026 non-GAAP gross margin to be 60% at the midpoint of our guidance range, as well as a substantial increase in our non-GAAP profitability. Infrastructure is now our largest revenue category and grew 145% year-on-year, driven by robust production ramps in optical data center-oriented platforms. Based on robust customer orders and rising visibility of program ramps, we are once again raising our expectations for 2026 optical data center revenue to be between $210 million to $230 million with continued growth as run rates expand into 2027. Keystone, our 100-gigabit per lane 5-nanometer CMOS PAM4 DSP and SerDes technology, continues to ramp into high-volume production at major hyperscale customers across the U.S. and Asia for 400-gig and 800-gig deployments, delivering almost 40% lower power consumption than competitors. Keystone's success serves as a foundation for multigenerational customer engagements that extend to the adoption of next-generation 1.6 terabit and 3.2 terabit optical scale-up and scale-out architectures at 200 gigabit and 400 gigabit per lane speeds, respectively. We expect Rushmore, our 1.6 terabit optical PAM4 DSP at 200 gigabit per lane speeds, to become an important optical connectivity growth driver beginning in 2027, which will layer on top of Keystone's successful ongoing ramp. Beyond our PAM4 DSP technology, we have comprehensively expanded our portfolio for optical and electrical scale-up and scale-out connectivity opportunities, including TIAs, drivers and onboard retimers. Together, these products support a broad range of data center architectures consisting of pluggable optics, LRO, LPO, NPO and CPO, providing customers with greater flexibility in their deployment of next-generation AI and cloud infrastructure using MaxLinear solutions. Washington, our stand-alone 200 gigabit per lane TIA platform, not only pairs seamlessly with Rushmore, but it can also be deployed stand-alone in LPO and NPO implementations that do not require a DSP. As the performance requirements for TIAs and drivers increase significantly at 200 and 400 gigabit per lane speeds, our deep SerDes expertise enables us to drive compelling performance advantages for such customer solutions. Annapurna, our 200 gigabit per lane Ethernet retimer platform, targets 1.6 terabit active electrical cable and onboard retimer requirements for scale-up in AI systems requiring low-latency short-reach electrical interconnects within server racks and switches. Annapurna's onboard retimer applications expand our presence into another critical layer of AI infrastructure. For both Annapurna and Washington, we expect initial revenue in 2027, followed by a more meaningful volume ramp in 2028. Outside of optical, our first XGS-PON hyperscaler design win for dedicated data center control train architectures has completed qualification for a 2027 ramp and beyond. Additionally, we have secured design wins for USB bridge controllers at two major hyperscalers for AI rack management alongside our broader analog and power management portfolio. These wins broaden our engagement across our data center platform and strengthen our strategic positioning with key customers. Our Panther family of storage accelerators addresses CPU, memory and storage bottlenecks. We expect revenues from Panther to roughly double this year with the potential to nearly double again in 2027. Outside the cloud data center, we expect edge AI-driven upgrades to 5G wireless WAN access and transport infrastructure to increase demand for our single-chip Sierra 5G radio SoC and our millimeter and microwave wireless backhaul RF and modem solutions in the mid- to long-term. Moving to broadband and connectivity: both categories delivered revenue growth in Q2, driven by large-scale deployments of our single-chip fiber PON and Wi-Fi 7 gateway platforms at major Tier 1 service providers in North America and Europe. We're also in the early stages of Ultra DOCSIS 3.1 and 4.0 deployments, which will provide additional stability to growth as ramps progress throughout 2027 and 2028. In summary, we are pleased with our first-half performance and the momentum we have in our data center business. Keystone has established MaxLinear as a proven high-volume, high-quality supplier of 400 gigabit and 800-gigabit PAM4 DSPs and SerDes technology. At the same time, our Rushmore, Washington and Annapurna active electrical cable and retimer platforms extend our reach into 1.6 terabit optical and next-generation AI infrastructure. With multiple revenue drivers layering on over the next two years, we believe MaxLinear is exceptionally well positioned for sustained transformative growth and increasing long-term shareholder value. With that, let me now turn the call over to Steven Litchfield, our Chief Financial Officer and Chief Corporate Strategy Officer. Steve?
Thanks, Kishore. Total revenue for the second quarter was $168.8 million, up 23% from $137.2 million in the previous quarter and up 55% from the $108.8 million in the second quarter of 2025. Infrastructure revenue for the second quarter of 2026 was approximately $85 million. Broadband revenue grew to approximately $45 million. Connectivity revenue was approximately $24 million, and industrial and multimarket revenue was approximately $15 million. GAAP and non-GAAP gross margin for the second quarter were 57.8% and 59.5% of revenue. The delta between GAAP and non-GAAP gross margin in the second quarter was primarily driven by $2.5 million of acquisition-related intangible asset amortization. Second quarter GAAP operating expenses were $101.8 million and non-GAAP operating expenses were $62.8 million. The delta between GAAP and non-GAAP operating expenses was primarily due to stock-based compensation and performance-based equity accruals of $36.5 million combined and acquisition-related costs and other costs of $2.2 million. GAAP loss from operations for Q2 was 2%, and non-GAAP income from operations in Q2 was 22% of net revenue. GAAP and non-GAAP interest and other expense during the quarter was $2.4 million and $2.3 million, respectively. GAAP EPS for Q2 2026 was $0.02 per share, marking a return to GAAP profitability. Non-GAAP EPS was $0.35 per share. In Q2, net cash flow provided by operating activities was approximately $4.8 million. We exited Q2 of 2026 with approximately $93.7 million in cash, cash equivalents and restricted cash. This included a substantial prepayment of wafers supporting rising demand for our data center products for which we have increasing order backlog in the second half of the year and into 2027. Our days sales outstanding in Q2 was approximately 28 days versus 27 days in the previous quarter, and our days of inventory was down in the quarter from 128 days to 123 days. This concludes the discussion of our Q2 financial results. With that, let's turn to the guidance for Q3 of 2026. We currently expect revenue in the third quarter of 2026 to be between $210 million and $220 million. Looking at Q3 by end market, we expect to see growth from all four of our business segments with particular strength in infrastructure driven by data center optical interconnects. We expect third quarter GAAP gross margin to be approximately 57% to 60% and non-GAAP gross margin to be in the range of 58.5% to 61.5% of revenue. We expect Q3 2026 GAAP operating expenses to be in the range of $98 million to $104 million. We expect Q3 non-GAAP operating expenses to be in the range of $66 million to $71 million. We expect our Q3 GAAP and interest and other expense to be in the range of approximately $3.8 million to $4.2 million. We expect our Q3 non-GAAP interest and other expense to be in the range of approximately $3.7 million to $4.1 million. We expect a $1.5 million tax provision on a GAAP basis and a non-GAAP tax provision of approximately $1 million. We expect our Q3 GAAP and non-GAAP diluted share count to be approximately 99 million each. In summary, our results this quarter reflect the continued strength of our optical products and the momentum we are seeing across multiple growth vectors within our infrastructure business. Our growth and innovation in this area has been transformational, and we believe we are in the early stages of a multiyear cycle characterized by revenue growth and expanding operating leverage. We're excited about the opportunities ahead and confident in our ability to create long-term shareholder value. With that, we'd like to open up the call for questions. Paul?
分析師問答
Our first question is from Tore Svanberg with Stifel.
Congratulations on the strong results. Kishore or Steve, you raised the optical transceiver revenue by more than $50 million for this year. Could you talk a little bit about what's driving that? Maybe talk a little bit about the regional nature of that? And also, if you could give us a sense for the mix between 400-gig and 800-gig.
Okay. Kishore, would you like to maybe take that one?
Yes. Tore, thank you. We are obviously very excited about the growing infrastructure business and especially about our success with our ramps for 400-gig and 800-gigabit optical PAM4 business. So as we entered the year, we were more concentrated in 400 gigabit revenues, but all the revenue growth we are seeing now is driven by 800 gigabit PAM4 success for us, and this will continue in 2027. And as we move later into 2027, our 1.6 terabit Rushmore will start to generate revenue and will drive growth beyond 2027 to 2028 and 2029. So yes, 800 gigabit is now substantially going to be a bigger portion of our run rate revenues moving forward. With regard to our end customers, we do not share particular customer names. Our customers span both the U.S. and Asia, hyperscalers and Tier 1 data center customers and OEMs. At this point in time, we are beginning to see more and more traction and revenue growth that will span both regions, including the United States.
Very good. And as my follow-up, just thinking about some of the expansion of products that you're now sampling, whether it's Washington and Annapurna, you talked about 2027 contribution. But I'm just curious, should the ramp be mainly in 2028? Or could you potentially start to see some ramps with TIAs and retimers already in 2027?
So our Rushmore product line, our Annapurna and even our TIAs for the 1.6 terabit or 200 gigabit per lane speed has been sampling now. It's in the customer qualification phases and design-in process. So we expect revenue to start generating in 2027. We expect the ramps at one or two opportunities to start somewhere in the second half of the year and then layer on top of that through 2028 and 2029. So yes, we do have some expectations of 1.6 terabit revenues and for the TIAs as a companion and for Annapurna active electrical cables in 2027.
Our next question is from Cody Acree with The Benchmark Company.
Congrats on the strong results and guidance. Maybe if I just get one point of clarification. The increase in the optical outlook, the $50 million increase, that is all just Keystone. Is that correct? That's not counting any Washington or Annapurna revenue in 2026?
That's correct, Cody. That's very correct. It's all driven by the Keystone product family.
And can you just talk about the visibility you have to that—the order visibility, the backlog builds that's looking into the second half? Can you just give us some color on the extension of your order trends?
Yes, Cody, I can answer that one. Across most of our businesses right now, the visibility is very good. Given some of the tight supply and the continued increase in demand, visibility is good. It's going out on or about six months anyway. So naturally, that gives us the confidence to go and raise these numbers.
And then just lastly, can you talk about wafer prepayments, maybe the amount that you did in Q2 and any expectation for Q3? And I guess, if you could just walk through some of the puts and takes for your gross margin improvement? Things like your mix and your incremental supply constraints and any kind of expedite fees that you may be paying?
Sure. Yes. No problem at all, Cody. Maybe just hit the prepayment real quick. Certainly, with this increased demand and making sure that we're securing wafers and products for our customers going out, we've certainly started to prepay in a lot of cases. So that was up a little bit in the quarter. We expect that to continue next quarter as well. But that's all against a product that has backlog out a couple of quarters, so we're comfortable with that. Regarding gross margins, we're a little ahead of schedule here and excited about hitting the 60% level in our guidance for Q3. The mix of our infrastructure products historically has been well above the corporate average. That continues to be the case today, and I see that continuing to expand over the next year or two as our infrastructure business grows and as our 800 gig and 1.6T products start to contribute further. We've been a little cautious about input cost increases—wafer cost increases, packaging and test increases that we're seeing in the industry. In some cases, you can pass these along to customers, so we've been a bit cautious on this front, but certainly we see improvements from here.
Steve, I just want to add: if you look at the revenue ramp we've had throughout 2026 and raised expectations, especially with the advanced nodes in 5-nanometer, Keystone is the only 5-nanometer SoC shipping in volume for the 100 gig per lane speeds—we were the first ones. We have gotten our forecast then kept going up. Our foundry and OSAT partners have been incredibly supportive in making sure that we can meet the surging demand as our qualifications went through, and we have started layering more and more customer product ramps on our optical products. So yes, supply is tight, but having strong relationships and constant communications with our foundry partners and OSATs has been incredibly helpful, and that goes a long way in meeting our demand.
Our next question is from Joe Quatrochi with Wells Fargo.
Maybe on the optical side—the revenue guide up—how should we think about the revenue run rate of that in the second half as we're just thinking about the trajectory into 2027?
Sure, Joe. As new qualifications come through and production ramps start, we began the year with a strong run rate and it's continued to improve. Raising this number here kind of sets expectations for 2027 as well. You would expect that there's not a stair step; as more customers' qualifications are completed and move into production volumes, those numbers go up, and I would expect that to continue into next year.
And then maybe on the broadband side, any update there in terms of what you're seeing from a demand perspective and some timing for these transitions. Has anything changed there?
Not a whole lot has changed. We've been gaining share on some of our PON programs, which has gone exceptionally well this year. Looking into the back half of this year and next year, telco CapEx continues to be good. Our customers continue to be rolling out new programs. As you recall, we've got content increases and a couple of other things, so everything is on track on that front.
Our next question is from Suji Desilva with ROTH Capital Partners.
Kishore, Steve, congratulations on the strong progress here. I know you're growing very strong in optical in 2026, but I'm trying to understand 2027 a little bit. Just what's the share opportunity—Kishore—as Rushmore upgrades happen from Keystone? Do you guys have an advantage there or perhaps even grow your share? Or should we expect that it holds from the success you have in Keystone?
Suji, the growth we are seeing comes from two factors: the market itself is growing very strongly, and we are gaining market share. Raising our expectation for 2026 revenue means a higher run rate into the second half of 2026, which has implications for 2027 as well. As we learn more about the ramps and how strong they are, we're upgrading revenue expectations. So it's happening in both ways: TAM growth and market share growth. Our performance differentiation and increasing traction with successful rollouts is accelerating ramps. Regarding Rushmore, Keystone is foundational for MaxLinear; it was the first major one that went to mass rollout from our standpoint, even though it represents our third generation of technology. Rushmore at 1.6 terabit is now sampling, with substantial performance and power advantages and unique supply chain diversification versus our competition. Considering all these factors, we feel very optimistic and excited about Rushmore and the upside potential of ASP increases with enhanced speeds. The same customers that are using Keystone are eagerly working towards deploying our 1.6 terabit. The qualification and interoperability cycle is a bit longer at higher speeds, but we feel well positioned to be successful with Rushmore as a successor to Keystone. Importantly, both 800 gigabit and 1.6 terabit will be workhorse speed nodes for a long time to come, so even as Rushmore ramps, Keystone will continue to be a substantial growth engine moving forward.
Okay. Layer on. My other question is on the TIA driver market, the Washington product and so forth. In the 1.6 terabit platforms, are you seeing more creative sets of CPO, LPO architectures that drive higher attach rates and make better use of your products as they break out some of the components there?
For the first time, we are positioning and marketing Washington as both a stand-alone TIA and paired with Rushmore. The first successes will likely be paired offerings with our own SerDes and PAM4 DSPs, such as Rushmore. As you go to higher speeds, our deep RF expertise is very valuable and differentiated and has potential to be used as a stand-alone product working with other DSPs and to be designed into LPO and LRO type applications. At this stage, we would expect first traction to come from pairing with our own device.
Our next question is from Quinn Bolton with Needham & Company.
Steve and Kishore, I'll offer my congratulations as well. I wanted to follow up on Suji's question just on Rushmore. As you look at the qualification programs you're engaged in now, is that an expanding set or expanding opportunity? Do you think you are going after more 1.6 terabit modules at your customers than you were originally looking at on 800 gig? Can you tell from the qualification activity whether you think your share continues to increase with Rushmore?
Very good question. From where we started with Keystone, we are now comprehensively designed across the board with the optical module players on 800 gigabit solutions. In a sense, 1.6 terabit now has to be designed to each of those customers where we laid the foundation with Keystone, having developed their modules, qualified them and interoped them. Keystone has created the footprint for us to roll out 1.6 terabit. Qualifying Rushmore is multifaceted: designing with module makers leads to the next phase of quals with data center operators, and that's when revenue ramps start. We expect this to happen toward the second half of next year with some initial revenue ramp starting in 2027.
Got it. And then Kishore, as you look at the broadband CPE gateway business and you talk to your customers, do those CPE boxes tend to use a fair amount of memory? Is the rising cost of memory causing any sort of delays in rollouts or perhaps lower units? Or do you think the CPE business is able to absorb the memory cost increases?
At this point, we have not seen memory cost be a major driver in our customers' decisions on using our products. In fact, our solution integrates a lot of functionality and on-chip memory, and many competing implementations use more external memory than we do. So we actually save our customers money due to our integrated solution. We have not seen any tangible impact on the volumes we were expecting and forecasting for this year and into next year. Lead times are long these days, so we get strong visibility based on backlog and bookings.
Our next question is from Tim Savageaux with Northland Capital Markets.
Congrats as well, especially on the guide. My first question: can you be more granular between overall market growth—unit volumes broadly speaking—versus share gain on MaxLinear's part, maybe at the expense of capacity-constrained competitors? I'd love to have you weight those two in terms of what's driving the step function here in the last couple of quarters.
Steve, go ahead.
I don't know that we can precisely break that out. From our perspective, we're confident we're seeing market share gains and the market is growing nicely. We're seeing our share increase, and part of the rationale for raising our guidance is that we've been able to take additional market share. We see that in the short term and expect to see it throughout next year as newer programs start to ramp.
Got it. And following up on the guidance: would it be fair to say the majority of that sequential growth is coming from optical in Q3?
I would say the majority is coming out of infrastructure. We're seeing growth across that end market much more so than some of the others. The others are going up, as guided, but a lot of the growth is coming from infrastructure, and we upped our optical guide, so that number goes up as well.
When you talk about growth across the rest of the segments, I assume you're referring to sequential growth there, not year-over-year?
Yes, just reflecting the guidance. That's correct, Tim.
Great. Last question: it looks like no 10% customers here. As you continue to ramp in optical, do you have the prospect of having one or more of these module guys as a 10% customer in the near future or in the future in general?
We've talked about being in a number of customers. Kishore mentioned it earlier as well. We've got a number of module makers and data centers that we're supplying product into today. Over time, yes, I do think you'll expect to see more concentration; I think that's well understood going into next year. I don't think that would be surprising.
Our next question is from Christopher Rolland with Susquehanna.
Congrats on the results, and I apologize if this has been asked as I joined late. But in terms of the composition, customer composition, particularly moving forward for DSP, is there any movement in terms of the balance between hyperscalers versus module makers and then also North America versus Eastern guys? Or is it still incredibly broad-based?
You might have missed this earlier, but we continue to see growth out of both regions. From a geography standpoint, we're seeing growth on both sides. As far as concentration itself, there is not a 10% customer today, but I expect, as we've talked about, you'll see a little more concentration over the next several quarters. There are only a handful of customers that will drive the most volumes over the next six quarters.
Okay. But just to be clear, you don't have one marquee customer pushing. Is that correct?
Correct, we don't have a 10% customer. It's a little broader based, but there are a handful of customers driving the most volumes.
Perfect. As a follow-up, can you provide any update on long-term metrics for the model more broadly? I think at one point you talked about 65% gross margins. Do you have an update on the longer-term model, including a path to 65%?
I don't think the target has changed. We feel like with the product mix and the end markets that we participate in, 65% is still the right goal, and I think there's a path to get there. The Q3 guide goes up a little ahead of schedule, reflecting our infrastructure business growing faster than other end markets, and those do have gross margins ahead of the corporate average. We are seeing increases in costs—wafer side, test and assembly—so we're doing our best to pass some of those costs along, but where we're paying premiums in some cases to meet customer demand, we're paying a little more right now. I certainly think there's a strong path to continued gross margin growth.
Our next question is from Ananda Baruah with Loop Capital Markets.
Same for me, I apologize if this has been asked—I jumped on late as well. As you look out the next couple of years and think about what the drivers of growth are, can you help us think about the order of magnitude—whether growth comes from bigger customer participation (i.e., hyperscalers) versus price lift from going to 1.6T and 3.2T versus broader growth in the marketplace?
Kishore, do you want to take that?
Ananda, it will take all of those factors: share growth, TAM unit growth and ASP growth as we move to higher speeds. Our footprint inside the data center is increasing as we offer a comprehensive product portfolio of TIAs, drivers, active electrical cables with Annapurna and onboard retimers. As the portfolio expands, it can address a number of architectures including CPOs, NPOs, LPOs and LROs for both optical and electrical scale-up and scale-out implementations. The offering has become more comprehensive, and we continue to expand the family to broaden our footprint. We feel we are making excellent progress getting these into the pipeline, which will result in multiyear revenue growth and expansion for MaxLinear.
That's really helpful, Kishore. Quick follow-up—anything notable either on the technical side or the relationship side over the last 90 days that is helping you move the ball forward that we should be aware of?
Every day matters. We're building on the successful penetration ramp happening with Keystone, and that is self-reinforcing. In the larger landscape today with the track record of millions of units of shipments of optical transceiver PAM4 DSPs, there are only three players, and we are one of them. That track record is very important. Having the next-generation offering with Rushmore and expanding the product portfolio all play roles in how relationships develop. We are now offering electrical and PAM4 optical solutions with TIAs, drivers, active electrical cables and onboard retimers. We also have storage accelerators in our portfolio that will become more important as agentic AI grows and storage bottlenecks become a limiting factor. Hardware acceleration and compression will be important to expand agents and reduce time to first token, which involves lower latency and improving power consumption. We are also increasing our XGS-PON for control plane in data centers. It will take many pieces to expand relationships with end customers. Our two big competitors are very large with many offerings, so it's going to take sustained work and expanding offerings to be a comprehensive player in data center infrastructure.
Our next question is from Karl Ackerman with BNP Paribas.
Two quick questions. Could you discuss some of the key drivers for your industrial and multimarket business into the second half and into 2027? This area has improved but is still halfway from the run rate you achieved in 2023, and it appears to be margin accretive. What are the drivers that could return it toward prior levels?
The industrial business has been recovering. Last year was very weak; we've started to see nice year-over-year improvements this year and I would expect that to continue next year. The industry itself is starting to recover, and we're seeing good pricing improvements in that region. I would expect pricing as well as new products to contribute to that growth.
If I may follow up, within broadband, could you discuss the mix of revenue on fiber today and whether you see fiber crossing over from cable broadband? Is that something that can occur in 2027?
Good question. We are a relatively new player in PON, but we've now won the top two players in North America. The second is ramping this year on track as we discussed. PON business is growing nicely, and I expect that to continue next year. It's hard to say when crossover will occur—I'd guess 2027 but it could push into 2028, because upgrades in the DOCSIS world are also growing. We're seeing decent growth on both sides and telcos are spending.
Our next question is from Tore Svanberg with Stifel.
I had a follow-up that's not been asked. Looking at the filing, it looks like your purchase obligations went up about $40 million, and another obligations item went up about $45 million. Can you explain a little bit the difference between those two? You talked about wafer prepayments and so on. Any more color on the difference is important.
The purchase obligations were the larger takeaway, and we did have some prepayments. With the stock price increase in the quarter, there were some payroll accruals that had to be done as well, including some stock compensation-related accruals. The majority of the increase is the prepayments for wafers supporting growth in Q4 and into Q1, as we start to place orders now for Q1. That's the majority of those commitments.
Got it. Last question: you were near 30% operating margin this quarter. How should we think about that number being a baseline going forward, especially relative to your OpEx guidance?
I don't want to guide beyond the current quarter, but our long-term goal is to be between 30% and 35% operating margin. You're seeing the model start to move in that direction pretty quickly. Profitability is good, and we expect growth next year. Gross margins are favorable relative to OpEx, though we'll see some OpEx increases to support growth. The operating leverage is compelling and we want to demonstrate long-term sustainability of these profit margins.
There are no further questions at this time. I'd like to hand the floor back over to Leslie Green for any closing remarks.
Thank you, Paul, and thank you for joining us on today's conference call. This quarter, we will be presenting at a number of financial conferences and virtual events. The details will be posted on the Investor Relations page of our website, and we look forward to speaking with you again soon. Thank you.
This concludes today's conference. You may disconnect your lines at this time. Thank you again for your participation.