ON - NASDAQ

Good day, and thank you for standing by. Welcome to the onsemi Second Quarter 2025 Earnings Conference Call. [Operator Instructions]. Please be advised today's conference is being recorded. I would now like to hand the conference over to your speaker today, Parag Agarwal, please go ahead.

Thank you, Kevin. Good morning, and thank you for joining onsemi's Second quarter of 2025 Results Conference Call. I'm joined today by Hassane El-Khoury, our President and CEO, and Thad Trent, our CFO. This call is being webcast on the Investor Relations section of our website at www.onsemi.com. A replay of this webcast, along with our second quarter earnings release, will be available on our website approximately 1 hour following this conference call and the recorded webcast will be available for approximately 30 days following this conference call. Additional information is posted on the Investor Relations section of our website. Our earnings release and this presentation include certain non-GAAP financial measures. Reconciliation of these non-GAAP financial measures to the most directly comparable GAAP financial measures and a discussion of certain limitations when using non-GAAP financial measures are included in our earnings release, which is posted separately on our website in the Investor Relations section. During the course of this conference call, we will make projections or other forward-looking statements regarding the future events or future financial performance of the company. We wish to caution that such statements are subject to risks and uncertainties that could cause actual events or results to differ materially from projections. Important factors that can affect our business including factors that could cause actual results to differ materially from our forward- looking statements are described in the most recent Form 10-K, from 10-Qs and other filings with the Securities and Exchange Commission and in our earnings release for the second quarter. Our estimates or other forward-looking statements might change, and the company assumes no obligation to update forward-looking statements to reflect actual results, change assumptions or other events that may occur except as required by law. Now let me hand it over to Hassane. Hassane?

Thank you, Parag. Good morning, and thank you all for joining us. In the second quarter, we made meaningful progress across our strategic priorities and advanced critical initiatives in Automotive, Industrial and AI Data Center. In Automotive, we're helping customers like Xiaomi improve range and enhance the driving experience and we're expanding our engagement with more global OEMs and Tier 1s, including our collaboration with Schaeffler. In AI Data Centers, we are enabling next-generation power architectures that drive efficiency and performance at scale through collaboration with market leaders like NVIDIA to accelerate the shift to 800-volt DC power architecture. We remain focused on making strategic investments to extend our competitive edge and deepen customer relationships to build long-term value. At the same time, we are making structural improvements across the business to enhance efficiency. This, combined with disciplined cost management, creates significant leverage in our model as we prepare to capitalize on a market recovery. On the financial side, we delivered Q2 revenue of $1.47 billion, exceeding the midpoint of our guidance and non-GAAP gross margin and EPS of 37.6% and $0.53, respectively. Turning to the demand environment. We are seeing signs of stabilization across our end markets. We have not seen any pull-ins to date due to tariffs and our diversified manufacturing footprint remains a competitive advantage, providing sourcing options to our customers as they work on optimizing their supply chains. By market, Automotive revenue in Q2 was down 4%, performing better than anticipated and is expected to grow in the third quarter with continued EV ramps. In China, select Xiaomi YU7 electric SUV models integrate our 1,200-volt EliteSiC M3e, enabling better performance and the longest range in this class. China remains a growth driver for onsemi with strong traction in both BEV and PHEV platforms. China revenue in Q2 grew 23% sequentially, driven by silicon carbide with the new EV ramps, I mentioned last quarter. We also expanded our collaboration with Schaeffler to deliver our next-generation traction inverter for a global OEMs PHEV platform using our latest EliteSiC M4T trench technology to unlock higher energy efficiency and reliability. We expect adoption to continue to expand and our customer engagement to continue to diversify as OEMs redesign hybrid architectures to meet the emissions target and extend range. Industrial revenue increased 2% quarter-over-quarter. Revenue for AI data center, which we report as part of our other bucket, nearly doubled again in Q2 over the same quarter last year. As outlined in the President's AI action plan, AI infrastructure has become a focus of national priority in the United States. AI growth will be limited by power delivery rather than compute alone and onsemi is the only broad-based U.S. power semiconductor supplier addressing this challenge with our intelligent power semiconductors, dramatically increasing power density and reducing energy loss. We are actively working with leading XPU providers on smart power stages that address the power requirements of current and next-generation platforms. We're in production on single SPS products in an industry standard 5x5 package and began sampling a dual SPS in the same footprint. As part of our ongoing transformation, we will continue investing in next-generation technologies where we have clear competitive advantages while reducing our exposure to areas with limited differentiation. This includes end- of-life of legacy products, exiting noncore businesses and repositioning our image sensing portfolio toward higher value segments such as ADAS and machine vision. These actions are reshaping onsemi into a company with a distinct value proposition, powered by leadership in intelligent power, sensing and analog mixed signal technologies. A strong example of the strategy in action is Treo. Momentum continues to build around our Treo platform with a design funnel that has more than doubled quarter-over-quarter as we progress towards our $1 billion revenue target. We are on track to doubling the number of products sampling from last year. Treo's differentiated technology, modular SoC-like design and ability to integrate high and low voltage domains are driving strong customer engagement across all our end markets. An example in Automotive is 10BASE- T1S, where we sampled over 10 customers as they work on their zonal architecture. After delivering our first Treo revenue in Q1, we've also reached an important milestone. We've now shipped over 5 million units from our East Fishkill facility this year. These milestones reflect the strength of our innovation engine and the strategic investments we've made to support long-term growth. By reducing complexity, sharpening our operational focus and allocating capital more efficiently, we are building a more resilient and higher-quality business for the long term. As the global economy accelerates to electrification and intelligent automation, next-generation vehicles, sustainable energy systems and AI data centers are converging around a shared need for a new era of power solutions. Beyond silicon carbide, our strategic investments in next-generation wide band gap semiconductors have delivered transformative gains in power density, thermal performance and energy efficiency. We started sampling customers on these new breakthrough technologies, and I will talk more about it soon. Let me now turn it over to Thad to give you more detail on our results and guidance for the third quarter.

Thanks, Hassane. Through our ongoing transformation, we remain dedicated to building sustainable long-term value for our shareholders. We have progressively rationalized our portfolio and manufacturing footprint to expand gross and operating margins at scale. These efforts will continue in future quarters, and we are committed to extracting value through our Fab Right initiative. Investments in next-generation technologies across the portfolio will continue to expand our position as a leader in power and sensing and drive the shift in our portfolio mix to move onsemi up the value chain with our customers. As a reminder, in Q1, we took aggressive action to reduce our manufacturing capacity and restructure our workforce to continue driving long-term operational efficiencies. In the second quarter, we began to see the benefits of those structural changes with a substantial reduction in operating expenses. In parallel, we increased our 2025 targeted share repurchase to 100% of free cash flow. We are executing to that target and after repurchasing an additional $300 million of shares in the second quarter, we have returned 107% of our free cash flow to shareholders on a year-to-date basis. Turning to the second quarter financial results. We exceeded the midpoint of our guidance with revenue of $1.47 billion, increasing 1.6% over Q1. Automotive revenue was $733 million, which decreased 4% sequentially, driven by weakness in America and Europe and offset by continued strength in China. Revenue for Industrial was $406 million, up 2% sequentially. While our medical and aerospace and defense businesses continue to grow, traditional industrial declined slightly in Q2 versus Q1. Outside of Auto and Industrial, our Other businesses increased 16% quarter-over-quarter with AI data center being one of the significant contributors. Looking at the quarter between -- the split between the business units. Revenue for the Power Solutions Group, or PSG, was $698 million, an increase of 8% quarter-over-quarter and a decrease of 16% year-over-year. Revenue for the Analog and Mixed-Signal Group, or AMG was $556 million, a decrease of 2% quarter-over-quarter and 14% year-over-year. Revenue for the Intelligent Sensing Group, or ISG, was $215 million, an 8% decrease quarter-over-quarter and 15% over the same quarter last year. Turning to gross margin in the second quarter. GAAP and non-GAAP gross margin was 37.6%, above the midpoint of our non-GAAP guidance. Manufacturing utilization was flat compared to Q1. Accounting for the capacity impairment completed in Q1, utilization is now 68% based on a reduced manufacturing capacity. We expect to see approximately $5 million reduction in depreciation on the income statement starting in Q4. Now let me give you some additional numbers for your models. GAAP operating expenses for the second quarter were $359 million as compared to $396 million in the second quarter of 2024. GAAP operating expenses decreased sequentially in Q1 -- as Q1 included restructuring charges of $539 million. Non-GAAP operating expenses were $298 million compared to $308 million in the quarter a year ago. Non-GAAP operating expenses decreased $17 million sequentially. And were above the midpoint of our guidance. This was due to delays in realizing the full benefit of our restructuring activities in the quarter, which we expect to recognize fully in the third quarter. GAAP operating margin for the quarter was 13.2% and non-GAAP operating margin was 17.3%. Our GAAP tax rate was 12.6% and non-GAAP tax rate was 16%. Looking forward, we expect no material change in 2025 due to the one big beautiful bill, while we see a positive impact in 2026 and beyond, reducing our non-GAAP tax rate to approximately 15% from our previous expectation of 19%. Diluted GAAP earnings per share for the second quarter was $0.41 as compared to $0.78 in the quarter a year ago. Non-GAAP earnings per share was $0.53 as compared to $0.96 in Q2 of 2024. GAAP and non-GAAP diluted share count was 415 million shares. Turning to the balance sheet. Cash and short-term investments was $2.8 billion with total liquidity of $4 billion, including $1.1 billion undrawn on our revolver. Cash from operations was $184 million, and free cash flow was $106 million. The sequential decline in free cash flow was driven by timing of working capital, which created lumpiness between quarters. Our year-to-date free cash flow is 19% of revenue, and we remain on track to deliver 25% free cash flow margin for the full year. Capital expenditures during Q2 were $78 million or 5% of revenue. Inventory was up quarter-over-quarter on a dollar basis by $9 million and decreased by 11 days to 208 days. This includes 87 days of bridge inventory to support fab transitions in silicon carbide, down from 100 days in Q1. Excluding the strategic builds, our base inventory is healthy at 121 days. Distribution inventory was 10.8% versus 10.1 weeks in Q1 and within our target range of 9 to 11 weeks. Looking forward, let me provide you the key elements of our non-GAAP guidance for the third quarter. As a reminder, today's press release contains a table detailing our GAAP and non-GAAP guidance. First, our guidance is inclusive of our current expectations that there is no material direct impact of tariffs announced as of today. Given our current visibility, we anticipate Q3 revenue will be in the range of $1.465 billion to $1.565 billion. Our non-GAAP gross margin is expected to be between 36.5% and 38.5%, which includes share-based compensation of $6 million. Our third quarter guidance includes 900 basis points of noncash under-absorption charges, and we expect utilization to be flat to up slightly in Q3. Moving on to non-GAAP operating expenses. We expect OpEx to be in the range of $280 million to $295 million, including share- based compensation of $32 million. We anticipate our non-GAAP other income to be a net benefit of $8 million with our interest income exceeding interest expense. We expect our non-GAAP tax rate to be approximately 16% and our non-GAAP diluted share count is expected to be approximately 410 million shares. This results in non-GAAP earnings per share to be in the range of $0.54 to $0.64. We expect capital expenditures in the range of $35 million to $50 million. As we look forward, we continue to rationalize our product portfolio to force the shift towards higher value and higher-margin products. In 2026, we expect that approximately 5% of our 2025 revenue will not repeat. This includes the end of life of certain legacy products, ongoing noncore exits and the repositioning of ISG that Hassane talked about. We've also been executing our Fab Right strategy to align capacity with this shift as we drive to a higher quality of revenue and long-term earnings power. To wrap up, we continue to operate with financial discipline and a clear focus on shareholder value. By taking decisive action to streamline our portfolio and align operations, we are well positioned for a recovery. We continue to invest in next-generation technologies and capabilities that will strengthen our competitive advantage and support our transformation. With our focus on intelligent power and sensing, we are reshaping onsemi into a more focused and differentiated company. With that, I'd like to turn the call back over to Kevin to open up the line for Q&A.

[Operator Instructions] Our first question comes from Ross Seymore with Deutsche Bank .

Hassane, the first one's for you. And I guess just kind of 2 parts to it. You sound better cyclically than you have in a while. So the first part is what are you seeing cyclically? And where are there still headwinds? Where are you seeing mainly the tailwinds? And perhaps more importantly, when we move to the secular side, you talked about the AI data center side, the Treo side of things, but we still have the offsets of businesses you're exiting, like Thad just mentioned probably a 5% headwind. Can you talk a little bit about the traction in those secular drivers and when the good ones are going to offset the exits?

Yes. Ross, thanks. That's a great segue into what really the call is about. So as far as what we're seeing, we're seeing stabilization relative to where we have been over, call it, the last 3, 4, 5 quarters, that is a positive development. We talked about Automotive hitting the low end in the second quarter. Thad talked about also expecting -- we expect Automotive to be up in the third quarter. So you're starting to see that stabilization. I'm not there calling a recovery. There's still a lot of uncertainty and customers are being cautious. But relatively speaking, I do have more, I guess, positively optimistic looking forward. But I remain cautious in the way we run the company until we see that stabilization turn into a better foundation for a recovery. We're controlling everything we can from what Thad mentioned from an OpEx perspective, we're being very disciplined in investments that will turn the company into the high-value products and revenue that we want. So those are the second part of your question. When we talk about AI data center, we've said we've started introducing products. Those products are gaining traction, not just from our analog mixed-signal, but also we talked about the silicon carbide JFET. So a lot of foundational technologies that we have in the company, we have been moving that into the AI data center and that business has doubled year-on-year from the quarter a year ago. That's the second quarter we doubled from last year as well. So that's getting the traction that we are expecting. Treo as a broad-based product, it is differentiated. We hit a very important milestone last quarter. I talked about we already recognized revenue a few quarters ahead of what we described prior to that. We expected the revenue in the second half of this year, but we posted the first revenue in Q1 of '25. And the second milestone is really the volume, and that tells you a little bit on the breadth and the traction that we're seeing with customers. All of these investments that we've been making over the last few years are what's driving our longer-term view and why it's important for us to start focusing on really reshaping the company into the high-value product company that we want and really a much better revenue quality from a margin perspective and from a capital allocation perspective. So all of these pieces of the puzzle that we've been investing in and putting in place, now they're starting to come together with traction in the market that we are able to use as a foundation for where we go next.

I guess as my follow-up, one for Thad, on the gross margin side, again, a little bit of a near-term long-term balance. In the near term, why is it flat to slightly down if your revenues are up? And then longer term, especially given the changes in mix that you're talking about, what are the key levers you think you can pull to get to the 53% long-term target, if indeed, that is still the target?

Yes. Ross, I mean, the key to margin expansion for us is all about utilization, right? It's -- we've been saying that for a few quarters here. As we see a recovery, utilization will improve, and that will fall through on a gross margin line a couple of quarters later as you think about burning through that inventory. In the short term, look, we're being cautious right now, right? We've got inventory on the balance sheet, we're going to burn through. We're lean there. We're lean in the distribution channel. We're in a good spot that when the market does turn, we take utilization up. But we're being cautious here. So we think it's flat to slightly up here in Q3 as we get better visibility into Q4 and to early next year, we'll think about taking utilization up, and that will give us a nice tailwind as you go into 2026. In terms of the march to the target of 53%, you've got about 900 basis points of underutilization charges in our Q3 guide. That's consistent with Q2. That, obviously, every point of utilization is 25 to 30 basis points of gross margin improvement. That math still holds. So as you see the utilization coming up, you'll see us get that 900 basis points back. We also have the monetization of the divested fabs as we moved that production back in-house. And we're continuing to do more work on our Fab Right initiative, which will give us another 200 basis points kind of in that neighborhood. And then as Hassane talked about, as we ramp these new products, they are at favorable margins. So I think when you start to add that up, you can start to get within kind of a pretty tight range of getting to that 53% long-term target. But like I said, in the short term, it's all about utilization. That's the #1 driver.

Our next question comes from Vivek Arya with Bank of America Securities.

For the first one, I just wanted to go back to Q2 results. So Industrial was a bit softer than I think you had thought before. So what drove that? And then the Other was much stronger. Is it really the data center part of that Other? And if that is the case, how large is the data center part of your Other business right now?

Yes. On the Industrial, it wasn't up as much as we were expecting. That's primarily because of what we call the traditional industrial. It was down just slightly. When we look at our traditional industrial, I would say it's kind of bouncing across the bottom. We think we've stabilized, but there's going to be some ups and downs here. It was down slightly and down more than we thought it was going to be. So that's the primary driver on the industrial. On the other market, yes, it's the AI data center and the opportunity that we have there. We talked about year-over-year that, that business has doubled. So still a small piece of the overall company, but growing nicely.

Okay. And for my follow-up, Hassane, where are we in the automotive recovery cycle for ON? I think you mentioned China appears to be strong for you. Is it really the weakness outside of China, especially at the North American EV OEM? Because I'm trying to contrast what you're seeing versus what several of your analog peers are seeing. Some of them are within, I think, 4%, 5% of their prior automotive peaks, whereas ON's auto business is still, I think, 30% of your prior peak. So why is the automotive recovery so slow for ON? And when do you think that your Auto business could start to regrow year-on-year?

Yes. So I think the Automotive specifically, the regions other than China, both Europe and North America are weak. I think there's a lot of uncertainty in the automotive market. I don't think we're any -- I guess, any different other than the portfolio rationalization that we've been doing, those obviously will not repeat moving forward. But where we are in the EV ramps continue to happen, of course, not at the same rate that we all expected. I think the unit volume is not where it needs to be. And the rest is just purely on the mix and exposure versus our peers. I can't really comment what our peers are seeing in automotive. But from our side, we hit the bottom in the second quarter. We're starting to post growth. Our expectation in Q3 will be growth. And then we'll see, based on the visibility we get over the next few quarters, where that's going to lead.

Our next question comes from Blayne Curtis with Jefferies.

I actually wanted to ask you about the ISG repositioning. And I thought you said the 5% was the end-of-life products, and that's what you've been saying for a while. So maybe you could just walk me through, I guess, what are you repositioning? Why? And is there a revenue number tied to that repositioning in ISG?

Yes. From a repositioning -- the strategic repositioning, I've talked about it at a high level in prior quarters is really our focus on the machine vision part of it. As we see some of the competition coming in, we've always said we're going to focus on value and we're going to focus on high-quality revenue based on the technology and the differentiation we bring. That differentiation locks with vision product or machine vision product rather than the human vision product. To give you an example that I've always -- that I've given in the past, reverse parking. It doesn't matter how good your camera is, there's always dirt on the lens because it's outside. The quality doesn't matter. That's where we're not going to be engaging on these designs. When it comes to proper ADAS, where the CPU or the SoC, the central SoC needs clarity and the best image quality for safety, that's where we bring -- that's where we add value with our vision products. That's the difference between where the company used to tackle, which is a lot of the volume aiming for #1 market share at -- across all markets to a very focused value-driven approach, which we've been on for a few years. That's really the difference that I talked about here. It's no different than the strategy we've been implementing. But right now, we're very confident in the approach that we want and the strategy we're going to move forward with and followed with the investments that we are making in order to maintain that differentiation in the markets we want.

And Blayne, to answer the second part of your question in terms of the revenue impact, we think for 2026, it's about $50 million to $100 million that doesn't repeat from the 2025 baseline.

[ Is that ] incremental to the 5% from the end of life stuff?

That's inclusive of 5%.

Got you. And then maybe, Thad, just on just the [ disti ] inventories, I guess, it kind of came in the high end of the range in June, kind of what's your expectation for [ disti ] in the September guide?

Yes. I think it's going to be right in this range, let's call it 10 weeks plus or minus, right? We came in at 10.8%. It's in our sweet spot of 9% to 11%. So we don't expect any material change one way or another.

And obviously, we talked -- so Blayne, just from a quarter-on-quarter, we don't really look at it quarter-on-quarter at that level of detail as long as it remains within the range because as you understand, there's ramps that we have in the third quarter. Those ramps kind of we get through in the second quarter, then you drain and you then get to a steady state. So that difference between quarter-on- quarter as long as it's within the range, for us, it's not something that we want to control at that level as long as we maintain a customer ramp strategy.

Our next question comes from Chris Danley with Citi.

Just a couple of questions digging on the gross margins. Is the 5% of business that's going away next year? Is that going to be gross margin accretive? And if so, how much? And then how does the silicon carbide business fit into the overall gross margin ledger? Are those gross margins lower than the corporate average now? And then how do we get those back to the 53% target?

Yes, Chris. So the silicon carbide gross margin today is below the corporate average, primarily because of underutilization. The key to getting those back to the corporate average is obviously volume and leveraging that manufacturing footprint that we have, which we will do over time as that business continues to ramp. In terms of the exits, long term, that is going to be dilutive to margins. Currently, it's somewhere around the corporate average. But when you think about our aspirations to get to a 50% plus gross margin, that business is not going to support that aspiration. So that's the reason to exit it now. We've said that -- and we've undercalled this for a few years here or maybe overcalled it, expecting to exit it faster. But long term, it will be dilutive. Short term, it will be neutral just given the fact that it's around the corporate average today. Now we will be rightsizing manufacturing as we go through this as well. That's what I said in my prepared remarks. So we are matching the Fab Right capacity with these planned exits.

Okay. And for my follow-up, just real quick. What percentage of your auto business is China? And then how would you expect that to trend over the next couple of years?

Yes. We don't -- look, we don't break Auto of China. We're disclosing Auto as a whole and China as a whole. We're not getting into that level of detail from the revenue cut. But we expect China to be a target market for us. We have a 50% share. That will continue to grow as revenue grows and as the number of units keeps growing. We're very happy with our position in China. Just to remind everybody, our focus in China is really where we add on the efficiency and the range. Every win that we have in China is tied to the quality and the performance of the products. And that's how we differentiate not just against some of our Western peers, but how we differentiate against some of the local peers. We will maintain that level of differentiation. I talked about we introduced our Trench silicon carbide already with some wins behind it. That is the way we're going to keep and stay ahead of everybody from a competition perspective. And that's the reason we're winning in China and really outside of China.

Our next question comes from Jim Schneider with Goldman Sachs.

I was wondering with respect to the Q3 guidance, you talked about Automotive being up in the quarter. I'm assuming that given your commentary, you're less sure about Industrial and Other being up. I was wondering if you could maybe give us a little bit of color on what you'd expect going into Q3 for there? I'm assuming that data center piece of Other would be much stronger. Maybe just kind of clarify where you expect to be auto -- excuse me, Industrial will be up, flat or down.

Yes. So the comment I made on Automotive is exactly that, but I made it specifically on Automotive relative to the second quarter being the kind of the trough as we ramp. But just to clarify my comments, we expect every end market, Auto, Industrial and Other to be up in the third quarter. Other will be up higher driven by the ramps that we see in these markets that we put under Other, which includes AI, of course.

Yes. And to give you a little more specific expectations there, we expect Auto and Industrial to be up low single-digit percentages. We think Other is going to exceed that will be up in the mid- to high single-digit range.

That's very helpful. And then maybe if you could give us any kind of color on within Auto, you talked about the U.S. and Europe being a little bit weaker. I don't think that's particularly surprising to anybody. But can you maybe talk about the reasons for that? Is it purely tariff-based uncertainty in terms of their end market uncertainty? Or do you think there's a little bit of excess inventory or buffer stock still trying to work down internally?

I would say it's all of the above. I mean I don't know what to point specifically at. It's not an industry or a market. Every customer has their pain points. Some have some inventory. We don't believe that's a broad statement from an inventory perspective. You have the tariff and you have just the general uncertainty of end market demand. So you see customers waiting to the last minute to place an order and [ an end ]. That's the cautious approach that we're taking. We've been more right than wrong in our approach. So we're going to continue to manage to the visibility into what we can see. But what you think about auto and -- or sorry, in general, Europe and North America is really all of the above that we all read in the headlines.

Our next question comes from Quinn Bolton with Needham & Company.

I just wanted to come back on that utilization impact. I think you said each point of utilization is 25 to 30 basis points. But you said there's kind of 100 basis points of underutilization charges included in the guidance. And so it kind of implies you need to get to 98% utilization to get that full 900 basis points. And I thought you guys in the past had said full utilization was more low to mid-80s. So can you just clarify kind of what -- where do you see full utilization?

Yes. Good question. So previously, we had said fully utilized for us was kind of in that mid-80% range. Post the impairment that we've done in Q1, that is now kind of in the low 90s, call it, somewhere around 92%, right? So if you do that math on the 25 to 30 basis points for every point of utilization, you get to somewhere around 700 points of improvement. There's also another 200 basis points of Fab Right initiatives that we're still taking. That will get you to that 900 between the 2 combinations. But yes, there's -- our expectation now is on the lower footprint, our fully utilized is now kind of in that low 90% range. So it's improved.

Got it. And then the rest sort of from the 46 or so percent to 53%, that would all be mix and new products?

Yes. Yes, exactly. Exactly.

And then...

Sorry, one more thing. As well as the fab divestitures, right? So you got another 200 basis points of the fab divestitures that we'll recognize.

Got it. Okay. And then Hassane, you'd mentioned the smart power stages, both single and dual phase that I think you said you're sampling now. Wondering if you could give us -- is that sort of about a year-long qualification timing? Do you think you could ramp faster? And maybe a similar question just on the 800-volt rack opportunity. Would you expect that to sort of ramp in the 2027 time frame given, I think, what NVIDIA has stated about its 800-volt rack time line?

Look, given the sensitivity with mentioning customers, whatever we have specifically on that opportunity is listed in the press release.

Okay. How about just the power stages then?

The power stages is really a standard design cycle. So we're expecting anywhere -- you can think about it as the 12 to 18 months qual cycle in production. Of course, we're -- that depends on end adoption as far as if there's any change in the road map. The most important thing is we're tied up with the -- on a road map specific with the XPU suppliers. So as they ramp and deploy their new platforms, we will be ramping with them. But obviously, we are in production on ours. So that -- the single SPS is ahead as we sample, our qual happens with the customer. So they're kind of different stages. But that's just to highlight, we have a road map, and we have a good cadence of new products now starting to get into that AI data center space across the whole power tree from high-power JFETs to really SPS close to the GPU or XPU. We've talked about it the last few years that we needed the new product engine to kick up. That's happening, and those are kind of the proof points that I'd like to highlight.

Our next question comes from Joshua Buchalter with TD Cowen.

I wanted to ask about inventory levels. I believe last quarter, you mentioned that you were expecting them to peak in 2Q and then decline through the rest of the year. Is that still the right way to think about inventories for the year? And any amount that you expect to take out through the balance of 2025? Like how should we think about utilization rates? I guess I'm a bit surprised to see them up when you're trying to bleed inventory given revenue was up modestly in the guidance for the third quarter.

Yes. Now keep in mind, the utilization is flat quarter-on-quarter. The calculation now is 68%. But if you normalize to pre-impairment, it's 60%, right? So it's flat quarter-on-quarter. The new calculation gets you to 68% utilized given that we have a smaller footprint and less capacity. First part of your question...

Inventory.

Yes. So inventory, yes, we're still on track here. We think we're peaking in inventory in Q2. We think it will be down slightly here in Q3. And I would expect in Q4, we continue to see that trend as we burn through that strategic inventory. That's always been our path. Our path is that we'll bridge -- that was bridge inventory for the fab transitions in silicon carbide, but we will be burning through that. So that's a nice tailwind to cash flow.

Okay. I appreciate the color there. And I'm sorry to keep picking on gross margins. But I wanted to ask about pricing. I think last quarter, you mentioned pricing a bit more aggressively to defend market share. How did that develop in the quarter? And any changes in the pricing environment that you've seen over the last 90 days?

Yes. No, no change to the pricing environment. It's actually stable, within our expectations. So there's nothing new here.

Our next question comes from Gary Mobley with Loop Capital.

I believe you said roughly a $300 million revenue headwind in fiscal year '26 or roughly 5% of revenue as you exit noncore business. How much of a headwind is it for fiscal year '25? And should we think about it as spread over 8 quarters, roughly $30 million to $40 million per quarter headwind? Or is it linear like that? Or is there any sort of [ step ] function?

Yes. For 2025, it's roughly -- we're expecting about $200 million of exits. Year-to-date, we're on track through Q2 of $100 million. Like I said earlier, we've continued to overcall this, we think we're going to exit it faster. So that pushes $100 million out into next year in terms of the exit. So that is the impact of 2025.

Appreciate that. I was hoping you can give us an update on the East Fishkill bring up sort of where you're at and looking forward what the impact to gross margin might be?

Yes. So from East Fishkill, obviously, it's the utilization impact overall that Thad mentioned from our fab network. East Fishkill is part of that fab network. So it comes with utilization. The important thing is we already have our power products, our silicon power products qualified and shipping from there, high voltage and medium to low voltage. Our image sensor qualification is on track as expected. And the Treo has started, I wouldn't call it a soft ramp at 5 million units already, but that's, again, a brand-new product that we ramped up with a brand-new technology in East Fishkill. So the fab is running and qualified where we want to qualify. And right now, it's primarily driven by utilization. But from a technology perspective, we're pretty happy with the performance.

Our next question comes from Vijay Rakesh from Mizuho.

Just a quick question on the Section 232, 301. Obviously, a lot of noise around that. How are you guys preparing for that? What are you expecting in terms of when this happens? And I have a follow-up.

I mean, I don't know, Vijay if anybody told you I have a crystal ball better than my peers. But I think the way we prepare for it is we remain focused on what we can control. We're talking about our footprint, our fab footprint, our manufacturing footprint as a competitive advantage. That has been recognized by customers, especially as the whole tariff talk has been for a few quarters now. The 232, I believe, will be very similar to the changes that we have to go through with customers. The thing is there's no planning to be done here because we don't know where it's going to land on either side, except the fact that we need to maintain flexibility, and we need to maintain the focus on what we can control.

Got it. And then in terms of silicon carbide, obviously, good to see you guys picking up some share there. One of your peers declared bankruptcy. Just wondering how that's playing out from a business risk perspective, obviously, people might be moving or reallocating. But just wondering how that's kind of playing out on the road map.

Yes. Look, from a road map perspective, that doesn't have an impact. I've always maintained my focus on we're going to win. We're going to win because of our own products and because of our own investment, not because one of our peers are struggling. So we're going to win because of things we are doing. Having said that, the changes in one of our peers with the bankruptcy, obviously, that is not the trigger that force customers to think otherwise. We all know they struggled way before the bankruptcy filing. I think a lot of road map changes from our customer, a lot of sourcing decisions have already been made. So that to me is all, I would say, part of the baseline, part of the funnel, part of the ramps. I wouldn't call the bankruptcy as a trigger.

Our next question comes from William Stein with Truist Securities.

I'd like to also dig into silicon carbide for a moment. I think one of the things that came out in the last quarter or so, Hassane, is that perhaps for some of your customers, there's been an interest in purchasing chips instead of modules. Modules had been the story. Now we're hearing more about Trench and some other maybe aspects of the individual chip design. Can you comment on that change? And if customers continue to choose chips over modules, does that influence your long-term thinking about the attractiveness of this market for you?

Yes. So that -- by the way, I guess, the shift or the change in mix between modules and what we call die is not new. That's been happening for a few years. That's already part of our baseline. As far as your reference to Trench, that's totally independent of die versus module. Trench is another step in device design. It was planar with our M3. Our planar performed better than everybody else's trench. We introduced and we have been winning with our new trench, which is our latest generation and we're winning because of the performance. Whether we put it in -- whether we sell it to a customer as die or we sell it to customers as a module, and we do provide both, still we win because of the performance of that die and we win more when we put it in our own module because we know how to design a module that fits our die specifically. A lot of our designs that are ramping, for example, even in China are a mix of die sales or module sales. We maintain the flexibility for what the customer supply chain wants to do and what they feel their core competencies are. Some customers have core competencies to make a module, some do not, and they prefer to buy our modules. We're here to offer flexibility. We've always said our intent is always to provide the most optimal solution to solve the customer problem and bring value. Value is in range of efficiency, and that comes on the die side with our trench technology. It does not change our view of the market. Obviously, from our results and the market share that we have been gaining and ramping, we've shown success no matter what the mix is, and we'll continue to do that. But the name of the game here is maintain R&D, maintain investment in key areas to provide differentiation, and that's the markets we're in.

If I can follow up in a similar area. One of the big consumers in silicon carbide has discussed a migration to hybrid inverter from silicon carbide to silicon carbide combined with IGBT. I think they said that they intend to reduce SiC by 75% in that traction inverter. But I think they haven't done it yet. I wonder if you're seeing that influence your go-forward view of this market, if it changes anything for you?

No change. This is not -- again, it's not the commentary from one customer. If you go back 2 years ago, in our Analyst Day, Simon already showed how the migration or the mix is going to be for higher end, higher range, higher performance will be silicon carbide. And as you go to more mainstream, you'll end up with a hybrid all the way down to there's still life in IGBT. So those commentaries are nothing new. I've addressed them multiple times. That doesn't change our view of the market. Our competitiveness in the market is to be able to provide a slew of technologies from silicon carbide, trench planar all the way to highly differentiated IGBTs because some vehicles still have IGBT on one axle, silicon carbide on another axle or full IGBT. It does not change the view of the market. We will continue to win based on the performance of the products specifically.

Our next question comes from Harsh Kumar with Piper Sandler.

Question for maybe Hassane. Hassane, as you think about the recovery and maybe the fuel for recovery, you talked about 900 basis points underutilization. Can I ask if there's an element of written-off inventory here as well that might come into play as you look at recovery?

No, no. When you talk about like reserve inventory or written off inventory, no, because we have a very disciplined approach on what you reserve and what inventory you write off. Inventory write-off does not have demand. So it cannot be part of the recovery. So when we talk about purely the 900 basis point being on utilization, it is purely demand driven that will drive utilization with healthy inventory that we build and ship to customers. If you recall, we saw the shift in market before a lot of our peers. So we took down our utilization ahead of most of our peers and ahead of the market really softening. So that puts us in a much better position from the inventory we have. We are already draining our inventory. And as demand picks up, we can pick up more utilization on the fab that will help the margin, help cash flow when we shift the bridge inventory. So everything we've done and the discipline we've shown all the way here is what's going to drive that healthy recovery without any kind of side notes.

Yes. And Harsh, we've got -- if you go back to our base inventory, take out the strategic, it's 121 days, right, in our sweet spot, right? So it's very healthy. We don't see an inventory risk on that and the strategic is just a matter of time of burning that through as demand comes back. Those are products that typically have long lives to them. So we don't see -- as we sit here today, we don't see a significant risk to any write-off of inventory.

Understood. And then as my follow-up, can I ask. You've got the legacy piece that you mentioned, I think, $50 million to $100 million, you'll peel off this year. You've also got some growth areas, which you called as other that are showing outsized growth. I was curious if you can help us understand what are the major components of the other outside of the AI data center piece and then maybe how big it is?

Yes. Look, the primary focus for the others bucket that I will talk about is really the AI data center. We have some client in computing, they're a very small part of our business. That's why we put it in Other. We're not breaking those out. As far as AI data center, it's still showing a lot of growth. It's showing per expectations because we're investing in that business. As we get more and more into that business and it becomes sizable, we may talk about it in more detail. But for now, we're just keeping it in Other, and that's really the driver for the growth you're seeing there.

Ladies and gentlemen, this does conclude the Q&A portion of today's conference. I will now turn the call back over to Hassane El- Khoury, President and CEO, for any closing remarks.

Thank you all for joining us. I want to take a moment to thank our global teams for their relentless execution, our customers for their continued partnership and our shareholders for their ongoing support. Together, we are building a more resilient and higher-quality business and I'm confident that the strategic progress we've made this quarter positions onsemi for long-term success. Thank you.