Hello, I am Ken Hsiang, the Head of Investor Relations at ASE Technology Holding. Welcome to our third quarter 2026 earnings release. I am joined today by Joseph Tung, our CFO. Thank you for joining us today. Please refer to our safe harbor notice on page two. All participants consent to having their voices and questions broadcast via participation in this event. If you do not consent, please refrain from asking questions or leave the session now. I would like to remind everyone that the presentation that follows may contain forward-looking statements. These forward-looking statements are subject to a high degree of risk, and our actual results may differ materially. For the purposes of this presentation, dollar figures are generally stated in New Taiwan dollars unless otherwise indicated. As a Taiwan-based company, our financial information is presented in accordance with Taiwan-IFRS.
Results presented using Taiwan-IFRS may differ materially from results using other accounting standards, including those separately presented by our subsidiaries. For today's presentation, I will give a brief business update and then go through the Q3 results. Joseph will close with our Q4 outlook. During the third quarter, our ATM capacities ran at or near full utilization rates. Demand for our services continues to outpace our ability to supply such services within our LEAP in much of our non-LEAP capacities. Our revenues grew faster than we had initially expected, driven primarily by faster than expected installations of high-demand LEAP capacities. Non-LEAP also performed slightly better than initial expectations. Even though historically speaking, third quarters tend to be driven by seasonal strength, given the increasing mix of LEAP services, seasonality related to underlying consumer buying patterns are becoming less of a factor in our overall loading patterns.
Overall, ATM profitability also improved more than anticipated, primarily as a result of higher than expected LEAP loading. Our EMS business's third quarter grew slightly ahead of our expectations. The third quarter continues to exhibit strong underlying seasonal demand for our SiP-related products. Our EMS business's profitability was in line with our expectations. Please turn to page three, where you will find our third quarter consolidated results. For Q3, fully diluted EPS was TWD 5.84, and basic EPS came in at TWD 6.13. Year-to-date through Q3, our fully diluted EPS was TWD 13.40, and basic EPS was TWD 14.17.
Consolidated net revenues reached a record TWD 244.2 billion, up 28% sequentially and 45% year-over-year. Gross profit was TWD 54 billion with a gross margin of 22.1%, up 1.1 percentage points sequentially and 5 percentage points year-over-year. The sequential improvement stems from higher LEAP content in our ATM business.
The annual improvement reflects higher ATM revenue and a more favorable NT dollar environment. Operating expenses totaled TWD 20.8 billion, up TWD 1.8 billion sequentially and TWD 5.1 billion year-over-year, primarily driven by higher headcount, profit-sharing expenses, and further R&D spend to support LEAP initiatives. As a percentage of revenue, operating expenses declined sequentially by 1.5 percentage points and annually by 0.8 percentage points to 8.5%. Our OpEx percentage declined principally from LEAP revenue growth after previous periods of R&D investment. We currently see operating expenses continuing to rise on an absolute level, but on a percentage basis should remain near the range demonstrated in the current year. This delivered an operating profit of TWD 33.1 billion, up 57% quarter- over- quarter and 151% year-over-year. We had an operating margin of 13.6%, expanding 2.5 percentage points sequentially and 5.8 percentage points annually.
Rising company profitability was primarily driven by higher loading and expanding LEAP product mix. Non-operating loss totaled TWD 0.4 billion. This amount includes net interest expenses of TWD 2.6 billion, up TWD 0.8 billion sequentially due to higher debt levels. Tax expense was TWD 4.8 billion. Our effective tax rate for the quarter was 14.8%. During the quarter, we saw beneficial changes to tax regulations and foreign currency movements, which positively affected our net deferred tax position. Given these changes, we now expect our full year effective tax rate to be between 16%-17%. Net income for the quarter was TWD 27 billion, up 28% sequentially and 148% year-over-year. On page four is a graphical view of our consolidated quarterly performance. For the third quarter 2026, our ATM business represented 62% of our consolidated holding company revenues, while representing 91% of our operating profits.
This is compared to 59% of consolidated holding company revenues, while representing 82% of operating profit in the third quarter of last year. As our ATM LEAP business continues to outgrow our other service areas, we continue to believe that our ATM business will continue to increase as a percentage of our overall business and profitability. On page five is our ATM P&L. For the third quarter of 2026, we recorded record revenues for our ATM business of TWD 151.5 billion, up TWD 25.3 billion sequentially and TWD 51.2 billion annually, representing an increase of 20% sequential and 51% annual growth. Gross profit for our ATM business was TWD 46.4 billion, up TWD 11.9 billion sequentially and up TWD 23.7 billion year-over-year.
ATM gross profit margin was 30.6%, up 3.3 percentage points sequentially and 8 percentage points annually, driven by higher operating leverage and a more favorable revenue mix, particularly a higher mix of LEAP. Our gross margin has surpassed our historical structural margin ceiling. A theoretical structural margin for our ATM business would now substantially depend upon the amount of expected LEAP product mix. Given that our LEAP business continues to grow rapidly, our margin range now appears to be dynamic rather than fixed structurally. During the third quarter, operating expenses were TWD 16.2 billion, up TWD 1.5 billion sequentially and TWD 4.3 billion year-over-year. The sequential and annual increases in operating expenses are primarily related to higher overall labor costs and general R&D expenses. Our operating expense ratio was 10.7%, down 1 percentage point sequentially and 1.1 percentage points annually.
We are approaching our near term ATM OpEx percentage target as we balance cost in a more technology-led growth framework. During the third quarter, operating profit was TWD 30.2 billion, representing a sequential 53% increase of TWD 10.5 billion and a 178% annual increase of TWD 19.4 billion. Operating margin was 20%, up 4.3 percentage points sequentially and up 9.2 percentage points year-over-year. The NT dollar had a + 0.5 sequential and a 3.3 annual percentage point impact to our gross and operating margins. On page six, you'll find a graphical representation of our ATM P&L. The chart currently highlights our gross profit margin. Over the displayed timeframe, our margin improvement has been largely driven by improving utilization rates within traditional product lines and increasing LEAP product mix. In this timeframe from peak to trough, our gross profit margin percentage has climbed 8.7 percentage points.
On page seven is our ATM revenue by communications, computing, and consumer applications. LEAP services are primarily included within our computing applications, with a lesser amount being included in our communications applications. As can be seen here, our computing applications percentage continues to grow steadily. With a large amount of our LEAP services coming from computing, revenues for our computing application grew by 39%. Meanwhile, automotive, industrial, and consumer all grew by about 14%. Our communications application had the least amount of growth during the quarter, growing 8%. At this time, we continue to believe that our computing application will lead growth into 2027 and 2028. On page eight, you will find our ATM revenue by service type. At the beginning of the year, we believed that our test business, led by rapid expansion of our wafer sort business, would outpace growth of our assembly business.
However, during the year, our assembly business has been showing even stronger than expected growth, especially as it relates to our traditional businesses. The demand for our test services continues to be robust. However, as with much of our business expansion, test capacities are awaiting new facilities. As stated during the previous quarter, we believe that both our assembly and test businesses will grow at similar rates during the year. On page nine, you can see the third quarter results of our EMS business. Our EMS business has historically focused more on products closer to consumer-driven end markets. Many of these products are experiencing a higher component cost environment driven by shortages caused by the rapid AI build out. These extra pass-through costs are increasing both our reported revenues and raw material costs. When netted, this ultimately negatively impacts our EMS business' margin performance.
EMS revenues grew 42% sequentially and 35% annually to TWD 93.5 billion. Sequentially, our EMS business' gross margin decreased by 0.6 percentage points to 8.3%. This change was principally the result of higher pass-through component costs and product mix differences. EMS operating expenses increased by TWD 0.4 billion sequentially and by TWD 0.8 billion annually. Our third quarter EMS operating expense ratio was 5%, declining 1.5 percentage points sequentially and 0.6 percentage points annually. Operating margin came in at 3.3%, improving 0.9 percentage points sequentially and declining 0.4 percentage points annually. The sequential margin improvement is the result of product mix offset by the higher component cost environment. The annual decline is the result of higher component costs. Our EMS third quarter operating profit was TWD 3.1 billion, up TWD 1.5 billion sequentially and up TWD 0.6 billion annually.
On the bottom of the page, you will find a graphical representation of our EMS revenue by application. Generally, the moves in the consumer and communication categories are related to the seasonality of the underlying products we service. Though the communications application percentage looks to have declined, on an absolute dollar basis, our EMS communications application grew quarterly. Annually, the communications application percentage declined primarily due to underlying customer production timing changes. The growth in the computing application is largely driven by business related to AI accelerator products. From a longer-term perspective, we continue to pursue potential synergies between our ATM and EMS businesses to co-develop system-level solutions. By integrating critical EMS competencies with ATM technologies, we have the potential to optimize entire systems end-to-end, applying the same co-engineering disciplines that successfully scaled SiP architecture.
As complexities increase at the die and package level, the system level will need to follow suit. On page 10, you will find key line items from our balance sheet. At the end of the quarter, we had cash equivalents, and current financial assets of TWD 107.9 billion. Our total interest-bearing debt increased by TWD 100.2 billion to TWD 406.5 billion. Total unused credit lines amounted to TWD 457.3 billion. Our EBITDA for the quarter was TWD 58 billion. Our net debt to equity ratio this quarter was 66%. As expressed in previous quarters, the third quarter represents the near-term peak of our net debt to equity ratio. We expect this ratio to come down somewhat in the fourth quarter. On page 11, you will find our equipment capital expenditures relative to our EBITDA.
Machine and equipment capital expenditures for the third quarter in U.S. dollars totaled $2.1 billion, of which $1 billion was used in packaging operations, $1.1 billion in testing operations, $41 million in EMS operations, and $5 million in interconnect materials operations and others. In addition to spending on machinery and equipment, during the quarter, we also spent $1.8 billion on facilities and land. We continue to believe that the growth of the company is highly dependent upon the execution of our various facility teams to deliver new manufacturing space and install equipment in those buildings for scaled-up mass production. We believe the ability to accelerate CapEx at this point to be beneficial to the company, as it allows us to bring in new manufacturing capacity at a time when we are running full.
Year-to-date, we have spent $8 billion of CapEx, $4.8 billion in equipment, and $3.2 billion in facilities. We currently have 22 ongoing building projects in Taiwan. We expect more than half of these projects will start supporting some level of revenue from now through 2027. One final point before handing the presentation off. Our overall business outlook continues to be increasingly defined by our leading-edge capacities. Though our LEAP services are frequently tied to AI-related products, we see our LEAP capacities as the manufacturing basis for future generations of products, not just AI-related. Multi-chip interconnectivity on leading-edge devices has become the norm. We are in the midst of an evolution of electronics production, where systemization is increasingly happening at the chip level. Simply put, what we are seeing is the next evolution of miniaturization.
It bears repeating, the capacities we are putting in place are not specifically made for AI. AI products just so happen to need the performance benefits LEAP interconnections provide. Our investments today are laying the future groundwork for generations of products to come. I'll hand the presentation over to Joseph to walk through the company's Q4 outlook.
Thank you, Ken. Let me give everybody a fourth quarter guidance. Based on our current business outlook and exchange rate assumption of $1 to TWD 31.9, which is that it remains the same as in the third quarter. Management projects overall performance for the fourth quarter of 2026 to be as follows. On consolidated basis, in NT dollar terms, our consolidated fourth quarter revenue should grow by 6%-8% quarter-over-quarter. Our consolidated fourth quarter gross margin should be between 23%-24%. Our consolidated fourth quarter operating margin should be between 14.5%-15.5%. For ATM business, in NT dollar terms, our ATM fourth quarter revenue should grow by 11%-13% quarter-over-quarter. Our ATM fourth quarter gross margin should increase by 100 to 200 basis points quarter-over-quarter.
For EMS, in NT dollar terms, our EMS fourth quarter revenue should be flattish quarter-over-quarter. Our EMS fourth quarter operating margin should be around 2%. Now, with that, we will open the floor for questions.
During the Q&A session that follows, we would appreciate it if your questions could be as clear and concise as possible and asked singularly. I, as the moderator, will be receiving each question and repeating each asked question. After an initial question, the participant may ask a follow-up question, clarifications of the earlier question, or another question entirely. Then we will move to the next participant. Participants may return to the queue for any additional questions and/or clarifications. Thank you.
If you have any questions, please raise your hand. Our first question is from Sunny Lin of UBS. Sunny?
Hello. Could you hear me okay?
Yes.
Thank you very much for taking my questions, and congrats on the very strong results. My first question is on your overall LEAP sales. Given the importance and your accelerating expansion, any update on your LEAP sales for 2027? How should we think about the split by full process, substrate outsourcing, and test?
For 2027, we are expecting to remain as we last projected to double our LEAP revenue for the year.
Well then, sorry, maybe just to follow up, that will be my second question. Any update on your LEAP sales for 2026?
Yeah. As I said, we are upping our 2026 LEAP revenue from originally $3.7 billion, that we mentioned last quarter, to over $4 billion for this year.
Got it. And--
Go ahead, please.
Sure. Sorry. Not sure if it is a connection issue, but I lost a part of the response. It is okay, we will check the transcript. My second question would be on the full process CoWoS. How is the overall ramp and yield for the key product second half of this year? Now, looking into 2027 with your expansion, should we assume a more diversified customer base and product base? Yeah.
Sunny, you are asking about our full-process business, and then looking for any update in terms of the full-process yield and such, right? Is that correct?
Yield, and then the product base going to 2027.
I think for 2026, our full-process revenue is on track, and we are keeping a $300 million revenue for the year. Everything else is going as planned. For next year, we are expecting over a $1 billion revenue coming from full process. Of course, the yield is at the level that we can get into mass production and major shipments.
All right. Thank you very much.
Does that answer your question, Sunny?
Yes. Thank you.
If you have any questions, please raise your hand. Our next question is from Gokul Hariharan of JPMorgan.
Yeah. Hi, Joseph and Ken. Thanks.
Hello, Gokul? Gokul, I cannot hear you. Okay, next question is from Haas Liu of BofA. Haas, can you hear me? Okay. Now we go back to Gokul Hariharan of JPMorgan.
Hello?
Hello, Gokul.
Yeah. Hi. Can you hear me?
Yes.
Oh, okay. All right. Thank you very much for taking the question. Hi, Joseph and Ken. My first question is on gross margin. Could you talk a little bit about where we are seeing this upside in gross margins compared to our guidance range for Q3? Looks like a very clean kind of 150-200 basis points beat and further upside in Q4 as well. Is it pricing or is it your full processes ramping up better in terms of margins? Secondly, on the same topic, I think now that we have already exceeded our previous structural range, and obviously the product mix is also different, you are still expecting LEAP to grow faster than the overall product mix. What is the kind of gross margin range we should think about as we think about the next couple of years?
Well, as we continue to ramp up our LEAP business, I think there will be further margin improvement in fourth quarter as well as going into next year. At this point, I think the so-called structural margin no longer applies because of the significant and rapidly changing business structure that we are having today. If we are to comment on our margins, I think the first thing we will say is that the so-called structural margin ceiling has now become the base for us.
Looking into next year, I think as we continue to ramp up our overall capacity, and also in a very friendly type of pricing environment, we believe that we will continue to see gross margin improvement on a quarterly basis from the base of 30%, and we will continue to see sequential improvement in our margin on a quarterly basis for next year as well.
Got it. Just to follow up there, if you look at the overall portfolio, maybe the LEAP testing aside within packaging, what would be the highest gross margin that you are able to earn in any of the LEAP processes? Is it already close to 40% for some of the processes, or that is too ambitious?
Gokul, you are looking for the potential highest performing margin for LEAP at this point? Is that your question?
Yeah. Within the three or four components that are there, which one is the highest performing, and what are those levels? I just want to understand what is the room for travel in terms of your overall gross margins, given your mix is clearly improving quite rapidly.
I don't think we want to comment on the different margins in different products or different packages. But I think overall, the overall LEAP business, both on assembly and test, our margin is accretive, and we will continue to see improvement in the margin expansion.
Okay. Got it. My second question is just a follow-up on your LEAP full process revenues for next year. Joseph, I think you mentioned more than $1 billion, but that sounds quite conservative given you are only doing $300 million in the first quarter of ramp this year. Could you talk a little bit more about your engagements on the full process? Also, if you ramp up all these 22 projects that you are currently ramping up in terms of facilities, both greenfield and brownfield, what does it mean in terms of, let us say, if you ramp it up in 2028, any indication in terms of what is the investment required from a CapEx perspective, and what does it mean for our continued growth? Does it mean that you continue to grow at similar rates even into 2028, not just for 2027?
Gokul, you are asking about a characterization of our full process business, and in certain regards, kind of a total return outlook. Is that correct?
Total growth outlook, given the very sharp ramp in the facilities.
We do have multiple customers engagement at this point, although we are pretty much constrained by our capacity, and we are aggressively building that up. From the business outlook that we have so far, I think our resources are quite limited. Although the demand is there, we are still trying very hard to catch up in terms of supply. For next year, we are expecting that the demand can be way higher than what we will be able to deliver. So right now, we are still projecting over $1 billion revenue coming from our full process. Anything else? I think the biggest bottleneck today is really still the space, although many different areas, we do have ourselves being constrained. But space is really one of the biggest constraints that we are facing today.
Okay. That is helpful. Thank you.
Our next question is from Laura Chen of Citigroup.
Hello. Hi, good afternoon, can you hear me?
Yes.
Yes. Hi, thank you for taking my questions. Just want to follow that $1 billion, the full process. Can you give us more color about that? What type of full process will see the contributions next year? Also, for our LEAP overall revenue. Obviously, still more driven by the substrate part. We do see that our companies have various different type of technology. Just wondering, can you give us more color about the growth breakdowns, and particularly in the full process in the next two years?
Hi, Laura, this is Ken.
Yeah.
I don't think we have a lot of characterization in terms of what types of full process. I think when we talk about full process, it's generally much more customer specific too.
Okay.
I think we probably can't really comment that much on that at this point.
Okay, sure. Understood. So that also seems like the growth into next year is more coming from the substrate part?
Yes. We are seeing very strong demand in test as well.
Sure.
Assembly and tests, we are both seeing a very strong momentum. Again, we are actually constrained with our capacity build-up. We are still trying very hard to try to catch up with it. In terms of full process, I think as I mentioned, we are moving ahead and on track. We hope that we can supply more for next year. At this point, we are conservatively looking at about over $1 billion revenue coming from that. There are some other new projects that we are entering. For example, CPO, and we are expecting over $0.5 billion of revenue coming from CPO as well for next year. That will be included in our LEAP revenue.
Oh, thank you very much. That is actually my second question about to ask about the CPO, because we see that Jensen earlier last week, he mentioned about that 30x the capacity will pick up for next year. I assume that also there is in the LEAP revenues. Will we going forward, to split that revenues breakdown for better investor understanding on that part growth?
I think we are breaking it down already in terms of oS, in terms of full process, in terms of wafer sort, and as I mentioned, CPO.
Okay, thank you. If I may also check on that, do we have any expectation on the CPO revenue contributions in two years?
In two to three years?
Yes.
No, we have a projection for next year, but we will have better clarity after we finish what we are setting out to do for now.
Okay, sure. Looking forward to that. Thank you very much.
Thank you.
Our next question is from Haas Liu of BofA.
Yes. Hi, can you hear me now? Sorry, I was disconnected.
Yes.
Okay, yes. Thank you so much for taking my questions, and congrats on the great results. I think just a quick follow-up on the structural gross margins, where you already reported the quarter's ATM sales, ATM gross margins up to slightly higher than 30%. So, with your LEAP business, that will be even higher in the next couple of quarters. Also, your mainstream packaging and testing business utilization pricing are actually also improving. Can we use third quarter gross margin as a base case going forward for your overall ATM gross margin assumptions as well as the cycle holes? Thank you.
Yes. As I mentioned earlier, that we are now using our structural gross margin ceiling as a base for next year onward.
Okay. So, we can use this kind of roughly 30% of the gross margins as the base case for next year. Got it.
Right.
Then--
We will see quarterly improvement for next year as well.
Okay. Sequential improvement. The second thing is just on CapEx. This year, you are already spending, based on what you reported just now, you have already been spending 80%-90% of your annual budget, which has been revised up twice year-to-date to $10.5 billion on CapEx. Would you be able to provide some update on that since there are still a few months to end this year? Looking into 2027, how should we think about that trajectory? As one of your foundry partner also just now announced to buy another panel fab. Should we also expect similar acquisition of the fabs or the facilities by ASE as well, and also more equipment procurement in the next couple of quarters to meet your customers' demand?
Yes. As I said, space continues to be one of the constraints that we have, and we are aggressively seeking for suitable existing plants for us to buy. But competition is pretty keen, and there are really not that many suitable factories for us to pursue at this point. So it's an ongoing effort that we are putting in. Also for this year's CapEx, we are increasing our CapEx for the year from $10.5 billion to $11.5 billion now. We're adding another $1 billion.
Yeah.
Of that $1 billion, half of it will be for facilities as we continue to fill up our new factories.
I see.
The other half will be for equipment, 60% of it will be for LEAP, and 40% will be for the mainstream events.
Got it. On the overall pricing environment across your LEAP versus the mainstream packaging and testing business, how should we think about since the supply-demand is still pretty tight?
Sorry to stop you there.
Okay.
You are on question number three.
Oh, sure.
[crosstalk]
Sure. Thank you so much.
Thank you very much.
Yeah. Okay. I'll be back in the queue.
All right. Thank you.
Our next question is from Gokul Hariharan of JPMorgan.
Hi. A couple of follow-ups from my side. Thanks for the opportunity. Joseph, I think you mentioned CapEx $ 11.5 billion, higher by $ 1 billion. Could you give an opinion on what you're thinking CapEx is going to look like next year? Are we still in an upward-sloping curve for CapEx? Could you also talk a little bit about, are you entering into any prepayment arrangements or anything like that with customers, given a lot of this new CapEx seems to be targeted at some of the key customers itself? Yeah, that's my first question.
Gokul, you are looking for an update on our CapEx outlook going forward.
Yeah, any funding arrangements that you have prepaid.
Yeah, the funding sources around it.
Well, we will continue to see very strong demand coming in, and we do have a very strong process for next year as well. At this point, I believe the CapEx spending for next year will be even larger than this year as we continue to ramp up our capacity. Also, in terms of R&D, we will continue to make the necessary investment to carry our missions. Funding-wise, we do have multiple cost-effective sources for us to raise funds to support our growth. I think, at this point, we will not be shy from making the necessary investment to support the expansion. I think for next year, we will continue to see negative cash flow. I think this thing will be balanced off in due course.
In terms of, as I mentioned, multiple funding sources, that will also include some of this trade financing that we can arrange with our different customers.
Understood. The second question is on the test business. Could you characterize the growth of the test business going into next year? Final test versus chip probing. I think, how constrained are you on test versus packaging? If you think about a relative scale, is the constraints on test bigger than packaging from a facility or equipment perspective?
Gokul, you're looking for an update in terms of looking at the test business going into next year, in terms of the growth rates relative to what we're seeing in terms of the market. Is that correct?
Yeah. Or related to packaging and related to constraints related to packaging as well.
I think the same pattern will likely in third and fourth quarter of this year, it will repeat into next year. I think, in terms of assembly, the growth rate should outpace tests, given the limited resource that we have. Also for tests, under the current resource constraints, we will need to prioritize our investment to more on wafer probe rather than final test.
Okay. Would you say test is more constrained in terms of your capacity versus demand compared to packaging, or it is largely similar?
I think it's mostly space. I think the space that we use for assembly is more efficient than tests at this point. So, given the constraint, we will put our priority on assembly, and for test will be for wafer sort.
Understood. That's very clear. Thank you.
Our next question is from Haas Liu of BofA.
Okay, yeah. Hi. Thanks for taking my question again. I think just a quick follow-up on the CapEx. A couple of your peers, including the one in Korea and also your foundry partner, mentioned that they are preparing for a significant or meaningful step-up on the CapEx for next year. Given this is the third time of CapEx upper revision for the year for ASE in 2026, do you also expect a similar statement to hold for your CapEx for next year? You already mentioned that next year's CapEx will be higher than this year, but just wondering whether that kind of aggressiveness will be applied to ASE as well.
Haas, you're looking for a characterization of how we view next year's investments.
Yeah. In a qualitative way as compared to your peers' comment for a meaningful step-up or just significantly higher than this year's CapEx. Do those statements also hold true for your CapEx outlook?
Haas, generally, we don't necessarily view our own CapEx relative to competitors or, I guess, peers in the market. In general, we do see that CapEx going into next year. We do see a lot of opportunity in terms of the business that will be available to us. As Joseph previously mentioned, we will not be shy in terms of investing to get at that business. Does that answer your question?
Okay, yeah. Just a related follow-up on the CapEx one is probably more specifically on the CPO that one of your key customers, last week in its Germany event, said it will hike CPO capacity by 15% next year at your facilities, then double again in 2028. Would you be able to share more detail on that? What is your role in CPO packaging and testing, what is the related CapEx you expect to spend, and how should we expect or quantify the related revenue contribution from that part of the business in the next few years from CPO based on your customer statement?
Haas, you're looking for a clarification in terms of the potential joint press release that we--
Yeah.
--had with our customer on CPO.
Yes.
Well, I don't think we will comment on specific product or specific customer. Only thing is, overall, I think we are the most active participant in the whole CPO effort and joining with our foundry partner as well as our customer. Just from the business that we are projecting to have from our customers, we're expecting to generate more than $0.5 billion worth of revenue coming from CPO for next year.
For next year. Okay. My second question would be just on the pricing environment across your LEAP business and also the mature or mainstream packaging and testing business. It seems that the supply chain is actually happening across the board. Will you be able to share some detail on how that pricing improvement has been helping your gross margins improvement in the third quarter? Into 2027, how should we expect that the price negotiation with your customers at this stage to help your gross margins expansion beyond the product improvement?
Well, I will not comment on pricing except we will continue to look for the most suitable pricing strategy depending on the situation. All I can say is overall, I think both for LEAP as well as for general market, I think we are in a very pricing-friendly environment at this point.
Okay. Could you also just try to quantify the--
Haas, I think you are on question number four.
Okay. I thought you just caught all of that. Okay. Yeah, I will be back in the queue .
Thank you.
Our next question is from Charlie Chan of Morgan Stanley.
Oh, hi. Good afternoon, Joseph, Ken, and [Iris]. Thanks for taking my question. Just a quick follow-up. Thanks for disclosing your CPO revenue. If we may, can we know roughly is the testing cost, for example, insertion three, insertion four, bigger revenue contribution or on-substrate a bigger contribution for that $0.5 billion revenue contribution?
I won't be able to comment on that, sorry.
Okay, no worries. Secondly, I think as you mentioned, you wouldn't hesitate to invest in space, capacity, whatever, to gain share, right? But recently, there seems to be some industry dynamic that one of TSMC's subsidiary, they're also getting into the on-substrate business. Do you think there is demand overflow or it is a real competition to your on-substrate business?
Charlie, you're asking for commentary in terms of competition within our LEAP business.
Yeah, because now it seems like demand is not an issue. My question is that who seems to have the space can really gain share, right? Is that a fair comment? Because one of the TSMC's subsidiaries ordering lots of equipments. We are a little bit worried whether there is a gain share at expense of ASE.
It is kind of sensitive to talk about a foundry partner's subsidiary situation. All I can say is, given the scale--
Right.
--I think we are really the dominant player in this field. Whatever competition, I think is out of necessity from our foundry partner as well as our customers.
Right.
At this point, I don't think it will move the needle that much in terms of the overall market share kind of situation. As I said, we are still in a catch-up mode, trying to supply as much as we can to meet the demand. I think that is all I can say at this point.
Got you. At least there wouldn't be any so-called price competition. It's like demand is too big, whoever can squeeze more space or capacity can gain share, but don't worry about pricing and margin.
I think the sheer scale difference is I don't think a price war will be created.
Makes sense. Thank you. If I may, non-AI. We start to pick up some negative points, PC and smartphone order cuts as a wafer front end. Except for iOS ecosystem. Joseph or Ken, can you share your observation about non-AI part of business and how that is going to impact your midterm outlook? Thank you.
Well, at least for this year, I think the general market growth is way ahead of our expectation in the earlier part of the year. I think we will have more than 20% growth in our general market business as well. We all understand that your cell phone and PC is going through a very soft period given the high component prices that is kind of hurting the overall demand. But that is more than compensated by the recovery of our automotive and industrial, and more importantly AI accelerator chips requirement. So we are seeing--
Thank you.
We are seeing very strong growth in this year for general market, and we are expecting that to happen again next year as well.
Okay. Thank you. Sorry for kind of asking questions so late. I was experiencing some technical issue, but thanks for all the answers. Thank you.
No problem. Thank you.
Thank you.
Next question is from Sunny Lin of UBS.
Hello. Thank you for taking my follow-ups. I have a question on very high level, what is your view on CPO versus NPO? Besides CPO, could ASE also capture the opportunity in NPO as well? Based on your commentary, will it be fair to say that based on the current visibility, it is fair to say that 2027 CPO will be able to ramp up, even with still some technical challenges or optimization that I understand you need to go through?
Sunny, you are looking for an update in terms of how we view the CPO versus NPO market, and then our capability to ramp up to support CPO going forward. Is that correct?
Yeah. Or maybe just CPO as a technology. Is it mature enough to start mass production for your key clients in 2027?
Yes. As I mentioned earlier on CPO, we are expecting mass production. We already started mass production in quarter four , and that will continue into 2027. As we are projecting at this point, we are looking at over half a billion worth of revenue coming from CPO. In terms of NPO and CPO, I think they will both coexist, and it is just a matter of difference in terms of overall performance between the two different types of packages.
And for NPO, will ASE be able to capture as well?
If there is real demand, yes, of course, we will be a major player in that.
Got it. Thank you. A quick question on EMS. Looking at the share, seasonality is a bit different. How should you think about the growth going to Q1? Typically, Q1 should be a down quarter as the build moderates. This time around, the flagship Sparkle clients are also launching for lower-end models going to early 2027. All things considered, how should we think about your EMS business maybe going to our first half of 2027?
Sunny, you are looking for the seasonal pattern of our EMS business, as potential underlying customer products change their timing of release. Is that correct?
Right. Thank you.
Well, we'll continue to expect for some growth going into 2027 as well for EMS business. I think the main focus will be on the high-performance computing. At this point, it represents roughly mid-teens of our overall EMS revenue, and we're seeing that ramping up pretty rapidly. Communication-wise or consumer-wise, at this point, relative to computing, it's weaker. All in all, I think our EMS will continue to move into optical, silicon photonics, and power-related type of devices. Also, of course, AI-related business that we'll be focusing on. Yes, we do expect further growth, particularly in the AI-related part of the business.
Got it. Thank you.
Thank you.
Next, we have Haas Liu of BofA coming back again. Haas?
Okay. Hi. Yes. Thanks. It is me again. I am just wondering if you could provide more clarity on the gross margins improvement third quarter. How much of that is coming from the product mix improvement versus the like-for-like price hike and the mainstream business? And your utilization improvements. Because if we look at the trend in the past eight quarters or seven quarters, that you already have experienced two step-ups on the gross margins, back in the second half of last year and now in the third quarter this year, there is another step up. As you accelerate your LEAP business, I think, going forward, in addition to the question I just had now on the third quarter gross margins breakdown from the relative benefits, I think just going forward, may I also check on your expectation whether that kind of gross margins improvement will further accelerate through 2027?
Haas, you are looking for further clarification in terms of what has caused our margin pickup during the third quarter, and then whether those types of improvements are sustainable going forward.
Yes. That is right.
Well, third quarter margin improvement is largely because of the LEAP ramp-up as well as overall improvement in utilization across the board, including the general market or mainstream packages and tests. Going forward, as we continue to ramp up our LEAP business and having a healthy growth on the general market business as well, I think, yes, we will continue to see sequential improvement in our margin on a quarterly basis going into next year.
Okay. I think just a follow-up on the application mix that in the third quarter, your computing business seeing a nice ramp. Would you be able to share more detail on that? What are the key applications are you targeting in that segment in the next one to two years between GPUs, CPUs, ASICs, and networking switches?
Haas, you're looking for an answer in terms of the composition of our compute business.
That's right.
Is that correct?
Yeah. And your targeted application in the next few years between GPUs, CPUs, ASICs, and networking switches.
I don't know if we can comment so specifically. Yeah. I think from a targeting perspective, we should be targeting everything. [inaudible]
Okay.
We can't be particularly specific. But in terms of our technologies, we do hope to be able to address all applicable products within that high-performance computing section.
Got it. Thank you very much.
Do you have another question, Haas?
No, I am good for now. Thank you.
Okay. I think you are the last question then.
Oh, okay.
Do we have any other questions online?
No. No more questions.
Thank you, everyone, for attending the call. Joseph, do you want to--
We had a fabulous third quarter, and we will continue to see strong momentum going to fourth and into next year. We will make the necessary investment to support the overall expansion, and we will continue to expect good results for the fourth quarter as well as for next year. Thank you very much.