Ben Poh: Good morning, ladies and gentlemen. I am Ben Poh, Head of Investor Relations, and today I will be moderating the call. On behalf of ASMPT Limited, welcome to our Q3 2026 investor conference call. Thank you all for your interest and continued support. Please note that all participants will be in listen-only mode during the presentation by the management. We will start the Q&A session after the presentation. During the Q&A session, priority will be given to covering analysts. Before we start, let me go through our disclaimer. Please note that there may be forward-looking statements about the company's business and finances during this call. Such forward-looking statements could involve known and unknown uncertainties and risks that could cause actual results, performance, and events to differ materially from those expressed or implied during this conference call.
Ben Poh: For your reference, the investor relations presentation on our recent results is available on our website. On today's call, we have the Group Chief Executive Officer, Mr. Bassel Haddad, joining us for the first time, and the Group Chief Financial Officer, Ms. Katie Xu. Bassel will cover the group's key highlights for Q3 2026 and provide outlook and guidance for the following quarter. Katie will provide details on the financial performance for the quarter. Now, I will hand the time over to our Group Chief Executive Officer, Bassel.
Bassel Haddad: Thanks, Ben. Good morning, everyone. Thank you for joining us today for our Q3 2026 earnings conference call.
Bassel Haddad: Let me start with a few opening thoughts. I've spent nearly three decades in technology, engineering, product, and business leadership roles across Fabless, IDM, and foundry environments. That experience has given me a broad perspective on how technology differentiation and value creation are evolving across the semiconductor industry. As the industry moves from scaling individual devices toward optimizing increasingly complex systems, the importance of integration, interconnect, and advanced manufacturing capabilities continues to grow. AI is a major accelerant of this architectural transition. As a result, advanced packaging is becoming a foundational platform for system-level innovation and differentiation. At the same time, the build-out of AI infrastructure is driving greater complexity and new opportunities across electronics manufacturing. This shift plays directly to ASMPT's strength in technology, application expertise, and customer engagement.
Bassel Haddad: The potential to position the company at the forefront of this transition is a major reason I chose to join ASMPT. My mandate from the board is clear: to build on our technology leadership and market position, and lead ASMPT into its next phase of sustainable growth, enhanced profitability, and long-term value creation. Two months in, I have engaged in numerous discussions with colleagues, met with customers across the globe, and spoken with shareholders and investors. My conviction has only strengthened. I've been particularly impressed by the depth of our technology and engineering capabilities, perhaps even more than I expected. Our opportunity now is to focus those capabilities where we can create the greatest differentiation, move faster, and translate our innovation into greater customer and business impact.
Bassel Haddad: Technology will be central to how I lead, but it must translate into impact, solving critical customer problems, scaling faster, and creating attractive returns. We will be technology-led, customer-driven, and financially disciplined. It is still early, and I am continuing to listen and learn, but I am very excited about the opportunity ahead for ASMPT. With that, let's turn to our results. Let me briefly share our business highlights for this solid quarter. First, we achieved record group bookings, driven by all-time high bookings for TCB, SMT, and Photonics. Next, in TCB Logic, orders for our C2S solutions reached a record high, driven by strong demand from AI data centers. Third, our Photonics business achieved record booking and revenue, driven by production ramps of 800G, 1.6T, and next-generation optical modules.
Bassel Haddad: Fourth, our group adjusted gross margin of 42.9% was up 35 basis points QoQ and 517 basis points year on year, reflecting improvements in both SMT and Semi. Finally, both our group revenue and adjusted EPS exceeded consensus. Now let me cover our business highlights in more color. Let's look at TCB, which achieved record booking in Q3. First, in Logic, we received bulk orders for C2S solutions driven by AI data centers. Specifically, this came from strong demand for larger compound die C2S from the OSAT providers of a leading advanced logic customer, as well as orders for other AI-related applications. In C2W, we secured repeat orders from a leading global IDM for advanced CPUs to support client computing and AI inference, reflecting this customer's continuous confidence in our C2W solutions.
Bassel Haddad: In memory, the industry is reassessing the optimal balance between memory density, thermal management, and cost, reflecting a recalibration of customer investment schedules. Thus, even though larger capability-driven investments are being pushed out, we believe the need for capacity additions and selective upgrades remains intact. Meanwhile, we continue to develop next-generation TCB solutions designed to deliver superior throughput, yield, and process performance, which positions us well for the next phase of HBM-related customer investments. Next, for TCB fan-out panel-level packaging, we are making good progress. Our quarter panel solution is currently undergoing sample bonding with a leading advanced logic customer, while we help develop solutions for a leading global IDM's embedded and surface silicon bridge die architectures. Our next key area of focus is Photonics. We have a comprehensive portfolio that spans pluggable optical transceivers, near-packaged optics, or NPO, and co-packaged optics, or CPO.
Bassel Haddad: This quarter, our Photonics business in Semi achieved record bookings and revenue driven by production ramps of 800G, 1.6T, and next-generation optical modules. Meanwhile, our pluggable optical transceiver solutions secured significant orders, reinforcing our leading position in the transceiver market. Notably, our Amicra business achieved a breakthrough NPO bulk order. In addition, Amicra's ultra-high precision bonding solutions received major orders from a leading communication services provider for optical transceiver solutions. In CPO, we continue to deepen our engagement with multiple leading global CPO players. Together, these developments demonstrate the breadth of our Photonics portfolio and create a multiyear growth opportunity. In hybrid bonding, our second-generation platform combines competitive alignment precision, bonding accuracy, footprint efficiency, and throughput. We continue to work with leading logic and memory customers on programs at various stages of sampling, evaluation, and qualification. Let's now cover the semi mainstream business.
Bassel Haddad: Revenue and bookings remain elevated, and it continues to be supported by AI-related demand for power management and server boards for data centers. AI infrastructure expansion also drove wire bonding and die bonding tool demand alongside high utilization in OSATs. Finally, SMT booking hit a record high for the third consecutive quarter and highest booking market share. Demand was driven largely by AI server boards and optical transceivers. SMT's high-flex solutions for large format board assembly continued to attract strong demand for AI server manufacturing. Notably, bookings growth in the Americas accelerated on the back of strong demand for AI servers. With these highlights, let me now hand the time over to Katie, who will walk you through our group and segment financial performance.
Katie Xu: Thank you, Bento. Good morning, everyone. Let me take you through the group's financial performance. I would like to reiterate that unless otherwise specified, the numbers I will be referring to today are for the group's continuing operations only, with adjustments made on the non-HKFRS measures. This slide covers our financial results for Q3 2026. In Q3, the group delivered a revenue of $669.2 million, up 6.3% QoQ, and 48.7% year on year, driven by Semi and SMT. Group revenue came in above market consensus. Group quarterly bookings reached a record high at $1.13 billion, driven by continued AI infrastructure investments, one-off bulk orders, and demand pulling due to extended lead times. Bookings increased by 24.7% QoQ, and 137.9% year on year, led by strong demand across multiple products, notably TCB, SMT products, and Photonics.
Katie Xu: Book-to-bill was 1.69, the highest since Q1 2021. Group adjusted gross margin was 42.9%. This was 35 basis points higher QoQ, and 517 basis points higher year on year, driven by improvements in both SMT and Semi. Group's adjusted operating profit was HKD 922.4 million, up 8.9% QoQ, and 379.2% year on year due to higher revenue and gross margin. However, adjusted opex was up 6.0% QoQ, and 16.6% year on year, mainly due to higher incentive provisions and increased volume. Adjusted net profit improved QoQ and year on year due to higher operating profit. Adjusted EPS was HKD 1.65, up 7.8% QoQ, and 489.3% year on year, which was above market consensus. Moving on to the Semiconductor Solutions segment.
Katie Xu: In Q3, Semi revenue grew to $374.2 million, up 1.5% QoQ, and 68.0% year on year. Year on year growth was largely driven by Photonics solutions serving AI-related applications, together with higher sales of wire and die bonders serving both AI and consumer end markets. Semi bookings were $563.3 million, up 31.3% QoQ, and 156.3% year on year, the highest quarterly booking level since Q1 2021. AI-driven demand benefited a broad range of Semi's advanced packaging portfolio, notably TCB, Photonics, and flip chip solutions. Semi Solutions adjusted gross margin reached 46.6%, up six basis points QoQ, and 502 basis points year on year. Year on year increase was largely due to high volume and a favorable product mix. Semi adjusted segment profit was HKD 595.4 million, down 1.3% QoQ, but up 620.9% year on year.
Katie Xu: QoQ decline was due to higher incentive provisions, while year on year increase was due to higher volume and better gross margin. Next, let me move on to SMT. SMT Q3 revenue was $295.0 million, up 13.2% QoQ and 29.8% year on year. Bookings reached a new record high of $564.7 million for the third consecutive quarter, up 18.9% QoQ and 122.1% year on year. This strong performance was driven largely by robust AI server demand and more than $100 million worth of bulk orders and pull-ins. Segment adjusted gross margin was 38.2%, up 131 basis points QoQ and 425 basis points year on year. QoQ and year on year improvements were driven by higher volume.
Katie Xu: SMT adjusted segment profit was HKD 410.4 million, up 44.1% QoQ and 143.2% year on year, driven by high volume. Now, let me hand the time back to Bento for the outlook and Q4 revenue guidance.
Bassel Haddad: Thanks, Katie. We expect Q4 2026 revenue to be in the range of $670 million and $730 million. At the midpoint of $700 million, this represents an increase of 4.6% QoQ and 37.5% year on year, and exceeds current market consensus. This will be driven by double-digit SMT revenue growth, while semi revenue is expected to be flat due to longer lead times of AP products. We expect Q4 bookings will remain significantly elevated year on year, supported by sustained AI infrastructure investment across both segments. However, Q4 booking is expected to be below Q3 record level due to the one-offs and pull-in in SMT. Looking ahead, AI demand is expected to drive sustained revenue growth in 2027. This concludes our third quarter 2026 presentation. Thank you, and we are now ready for Q&A.
Bassel Haddad: Let me pass the time back to Ben to facilitate.
Ben Poh: Thank you, Bassel. Ladies and gentlemen, we now begin the Q&A session. To ask a question, please click raise hand on Zoom and I will request you to unmute. Please limit yourself to two questions each time. First, I would like to request Gokul to unmute to ask the question.
[Analyst 1]: Yeah. Hi. Good morning. Thanks, Ben, and welcome, Bassel, to this inaugural investor call that you're having. Maybe first question to you, Bassel, given that you've spent close on three months at the company. When you look at the portfolio of ASMPT and you think about moving forward into the strong growth in advanced packaging or advanced backend itself, could you talk a little bit about where you are seeing potential areas that you could augment the portfolio further? What is the strategy to get there within the overall advanced packaging space? Also, given previously ASMPT has indicated the inclination to potentially divest SMT business, any updated thoughts on that thinking itself, given SMT has seen a pretty good resurgence in terms of business momentum in the last quarters? Yeah, maybe that's the first question.
Bassel Haddad: Okay. I'll try to answer both sides of them. The first one was about my three months into the company. Not yet, actually. Two months and a half. Let me start with that. So I'm two months in, and still in deep listening and learning mode with our technology teams, customers, and markets. As I shared in my comments or prepared remarks, now I'm very convinced, as I was before, even more, that we have a strong technology, engineering depth and customer relationship, and that's across advanced packaging, Photonics, and electronic manufacturing. As I also shared, we're going to be technology led and drive where we see the biggest return on our investment. So the foundations are very strong.
Bassel Haddad: I'm energized with what I see, and we're going to just invest where we see the market growing and where we see our set of IPs and portfolio and technology delivers the best results. As I move to the second question about SMT, I cannot disclose the details at the moment of where we are in the process. We will certainly update the market when we can. With that said, SMT records booking clearly reinforce and reflect the quality and the strategic value of that business. The assessment remains focused on maximizing long-term shareholder value.
[Analyst 1]: Got it. Maybe one question, second question is on the TCB side of the equation. I think you did talk a little bit about HBM. There is some pause on HBM for qualification and potentially some density related concerns, which I think the market has also talked about potential beat backing in HBM, et cetera. Previously, I think we have thought about memory TCB or the memory opportunity in TCB being much bigger than the logic opportunity in TCB. Now that logic is growing quite nicely, maybe you are starting to get some breakthroughs into chip to wafer SML. Any updated view on the related size of these two markets within TCB memory versus logic? Does this memory push-out on HBM orders change your longer term addressable market expectation for TCB at all? I think the last one we have is $1.6 billion by 2028.
Bassel Haddad: Thank you, Gokul. Yes, as we mentioned, we believe what we see in HBM layer count stagnation reflects the industry assessment of the optimal balance between memory density, bandwidth, and thermal. Even though we see larger investments have been pushed out, we believe there will be the need to add capacity and selective upgrades. All in all, it is an adjustment to the timing rather than a change in the underlying demand. Just to summarize the HBM side of things, we continue to see long-term potential for HBM4 12-high and beyond, alongside HBM 8-high, both serving different segments of the AI market, and we are well positioned in both areas. To address the other part of your question, in terms of that TCB TAM, we expect the TCB TAM to expand beyond $1.6 billion by 2028.
Bassel Haddad: The expansion will come from different areas, including logic across CPU panel and C2W. We will provide the official TCB update one year and only by the year-end earning call in Q1 2027. However, with that said, we do see that the split or the majority of the TCB TAM in the next couple of years would come from logic versus memory, versus HBM.
[Analyst 1]: Okay. That is very clear. Thank you very much. I will go back to the queue.
Ben Poh: Thank you. Thank you, Kok Ho. Next, I would like to request Sunny to unmute to ask question.
[Analyst 2]: Sure. Thank you very much for taking my question. Good morning. I also want to follow up on the addressable market for TCB. Basil, may I clarify, just now you mentioned the addressable market should be larger than $1.6 billion by 2028 if we also consider upside coming from silicon bridge integration, also panel level?
Bassel Haddad: Yes, that is correct, Sunny.
[Analyst 2]: I see. Well, even that estimate was provided by I think March 2026. Fast-forward, now we are indeed seeing progress being made for silicon bridge integration and also panel level. Understand, you may not be able to update this time around, but should we expect pretty meaningful upgrade to that addressable market that you anticipate a few quarters ago? I also want to understand, especially for silicon bridge integration, could you share with us your current progress in terms of timeline for qualification and also potential volume shipment? Also for this specific TCB tool, how should we think about the tool price and also gross margin, compared with the typical TCB that is shipped to the larger foundries or relevant OSAT providers?
Bassel Haddad: Okay. Let me try to address the different parts of your question. On the first one about the TCB market share, you are right. As I said, we expect the logic portion of the TCB TAM to be the majority of that TAM in the next few years. It would come from things like silicon bridge and fan-out panel-level packaging as well as C2W demand, Sunny. Again, the numbers and the color will be provided in our year-end earning call. Can you remind me of the second question, second part?
[Analyst 2]: Yeah, I want to know a bit more details around your progress for silicon bridge integration type of TCB. So, the timeline for qualification, potential volume shipment, and also the tool price and gross margin for this specific opportunity.
Bassel Haddad: Okay. Thank you. Our quarter panel solution has progressed to sample bonding with a leading advanced logic customer. We are also developing solutions for a leading global IDM embedded and surface silicon bridge die architectures. Basically, the timing of it is dependent on the ramp of the customer, and we will be there when the ramp happens. In terms of the gross margin and the pricing, I would say it is too early to talk into it, but we see that falling into the same range of our advanced packaging portfolio.
[Analyst 2]: Got it. Thank you very much. My second question would be to look at silicon photonics. I would like to have your view on the energy upgrade to NPO compared with CPO. How should we think about the pace of ramp between these two solutions? How should we view ASMPT's position in NPO compared with CPO in terms of capability, competitive positioning, et cetera?
Bassel Haddad: Yeah. As we mentioned, we have a very comprehensive portfolio that spans pluggables, NPO, and CPO. We are working with our customer base in the different regions to cater to their demands. We believe we have a unique product portfolio, and we mentioned Amicra, that is gaining very good traction in the market. So we are well-positioned across all of these sub-segments of the Photonics market, Sunny.
[Analyst 2]: Well, so, I think earlier, management mentioned that CPO maybe for ASMPT would be more of a driver into 2028, if not earlier. So I wonder if that is still the case, based on the current visibility. Do you think NPO may potentially ramp a bit earlier compared with CPO for ASMPT?
Bassel Haddad: Yeah. I think what we will try to do is to provide you a view of the Photonics market again along the same time with our TCB market TAM update. We expect NPO to ramp earlier than CPO. I think we look into CPO from a volume production more towards the end of this decade. All in all, as you have probably seen the results, 2026 will be a high-growth year for Photonics. We still expect Photonics to grow significantly in H2 compared to H1. We also expect growth in 2027 coming from these different segments of the Photonics market.
[Analyst 2]: Got it. Sounds good. Thank you very much.
Ben Poh: Thank you, Sunny, for your question. Next I would like to request Kevin to unmute to raise question.
[Analyst 3]: Hey.
Ben Poh: Kevin, please.
[Analyst 3]: Yes, sure. Thank you, management, for taking my question. My first question will be more on the order outlook side. I think we are seeing a very strong booking to stay elevated next quarter. At the same time, we are also seeing the semi revenue to be largely flat. You mentioned that this is due to a longer lead time for an advanced packaging product. I am just wondering approximately what percentage of this revenue is coming from advanced packaging? I believe last time we did mention also that some of the components' lead time is getting longer. I was just wondering, are we seeing any kind of supply bottleneck or capacity constraint due to the change in product mix or the component supply?
Bassel Haddad: Okay. Thanks for your question, Kevin. Let me try to address them. From a capacity, we have enough capacity to meet demand, even with the increased booking we are seeing. We are not concerned in that regard. In terms of the lead time, we have provided that guidance of six to nine months in AP. That continues to be the case. We are working diligently with our suppliers to ensure that we get what we need from their components and material that we can ship on time. You are right in the fact that you see the booking kind of running a bit ahead of billing. That is because of the AP mix becoming bigger part of booking, and that realization into billing is happening over two to three quarters, six to nine months. Did I miss other parts of your question? Anything else?
Ben Poh: Kevin, could you go ahead with the second question?
Ben Poh: Kevin, are you still there? Kevin?
[Analyst 3]: Oh, hello. Yeah.
Ben Poh: Yeah.
[Analyst 3]: Sorry, I was muted.
Ben Poh: No.
[Analyst 3]: Yeah. Thank you. The second question I have is, I was wondering, can we share a little bit more on the eMP TC business potentials? Like what kind of process step are we currently involved, and potentially what can we expect the timing of this potential revenue?
Bassel Haddad: Yeah, as I shared before, we are working with our customer on that. We are engaged with them in all the right process development, qualification steps. The timing of it is really, Kevin, a function of the ramp of the technology by the customer. What we are basically looking for is to be there when the ramp happens and serve our customer needs with our products. We believe we have a very competitive product for that market.
[Analyst 3]: Okay, great. Thank you. It is clear.
Ben Poh: Thank you. Thank you, Kevin. Next, I would like to request Daisy. Daisy to unmute to raise question.
[Analyst 4]: Oh, hi. My first question is about HBM use TCB, because some market intelligence firm expect that 8-high is project to see sharp rise induction from mid-2027 onwards, overtaking 12-high configuration. I remember, if I remember correctly, that your HBM use TCB started from 12-high back in October 2024. Do you see any business opportunity for 8-high HBM? Also I want to ask, how do you see your HBM use TCB's China opportunity? Currently HBM is a bottleneck for China AI GPU shipment.
Bassel Haddad: Okay. As we mentioned before, as you have seen in industry reports, there is kind of a stagnation in the HBM layer count. We believe TCB is needed in some architectures for HBM 8-layer. We continue to work with our customer in this side of the industry to ensure that we have the right product, and we do the qualification with them. When the ramp of the demand for TCB and HBM happens, we will be there, and we are excited about the products we have shipped and also working on that would be available for them to use when that capacity as well as capability build happens in the HBM market. Towards the China side, we do not comment specifically about customer. We are a global business. We continue to serve our customers across all major subregions of the world.
Bassel Haddad: We are engaging with our customer to see how our products can help them.
[Analyst 2]: Thank you. My second question is for Katie so how should we model the share-based payment? Because 2026 is a very good year for ASMPT, and the next year, the management expect a sustained revenue growth. Will the share-based payment also increase next year?
Katie Xu: You are talking about the incentives, right?
[Analyst 2]: Yeah, incentives.
Katie Xu: Yeah. Share-based incentives actually for us is non-GAAP item. Just want to call this out. But overall, yes, you are correct. As we continue to see sustained revenue growth, we do anticipate that the share-based payment will increase somewhat. But I do not have the details, and I will not comment on exact, but I think directionally you are correct.
[Analyst 2]: Thank you, Katie.
Ben Poh: Thank you, Daisy. Next, I would like to request Arthur to unmute to raise a question. Arthur, please.
[Analyst 5]: Hi. Yes. Thank you. Welcome, Ben. Last time we met in SEMICON Taiwan. Happy to see you again. I have two questions. One to the overall business, and second is to the gross margin. Maybe I will ask CFO. The first question on the overall business. You mentioned that the lead time of the component is getting longer and longer. What else management team can do to shorten the lead time and drive the delivery? That is my first question. Thank you.
Bassel Haddad: Yeah. Thank you, Arthur, and great to meet you again here. Just to be clear, based on what we have said a quarter ago, the lead time has not increased. We are in the same for AP, six to nine months lead time range. We are aware of the industry supply situation, and we are working very diligently to ensure that we get the components on time to deliver to our customer schedules. That includes things that we do in terms of safety stocks and risk buys for some components that we deem could be the bottleneck in our delivery time. So far, this is going per the plan, and we do not see any change to the six to nine months delivery and lead time we mentioned before.
[Analyst 5]: Okay, thank you. Second question on the gross margin. Hi, Katie. I recall that in the SMT history, sometimes the margin was driven by two factors. One is product mix, one is scale. The gross margin for the customer, either transfer, actually was lower, and AI or the auto actually is higher. Can you give us some color for the upcoming quarter and also the SMT business trend? Thank you.
Katie Xu: Thanks, Arthur. For SMT gross margin, maybe I should comment really quick on Q3, like how it played out. You see that the Q3 gross margin for SMT has showed quite a bit of growth QoQ and year-on-year, and it is mainly driven by volume. Arthur, this is the point that you are making about scale. You see that the SMT actually has, the gross margin actually has pretty good volume leverage. Going forward, we do expect the SMT gross margin to stay at this level, since the volume will be staying at this relatively high level. In terms of your comment on the product mix, you are correct. Auto industrial usually could command a relatively better margin. But going forward, I think other than auto industrial end market, Arthur mentioned already, SMT business is driven by AI demand as well.
Katie Xu: We do see opportunity there, to make sure that the gross margin, again, at this revenue level, can sustain. Thank you.
[Analyst 5]: Okay. Just to clarify, AI SMT could also improve the mix in the future?
Katie Xu: The revenue mix will be coming from AI demand.
[Analyst 5]: Okay. Got you. Okay. Thank you.
Ben Poh: Thank you, Arthur. Next I would like to request Leping to unmute and raise question.
[Analyst 6]: Okay. Thank you for taking my question. I have two questions. The first question is about your Photonics business. You mentioned this, you have Amicra is adopted by a major global communication service provider. So how important is this adoption? Does it mean that your equipment is, or the NPO is widely adopted by this customer and that you will see a meaningful adoption in the future? Since typically you do not mention service provider. Typically, I believe your equipment is mainly sold to OSAT or the IDM companies. Thank you.
Bassel Haddad: Oh, thank you. Thank you, Leping. So yeah, as you would expect, we cannot go into the specifics of the customer. With that said, I would say that we expect this opportunity not to be one time order, but continuous, as we continue to deliver and our product meets the customer needs. So we expect continuation of this customer engagement throughout 2027.
[Analyst 6]: Okay. Thank you. The second question is about the panel level packaging. What is your view on the customer's adoption, the speed of adoption on this technology? Currently, are we still in a trial phase, and will you expect the high volume manufacturing or the high volume shipment of your panel level packaging equipment? Thank you.
Bassel Haddad: Yeah. I think I touched on it before. Panel delivers good benefits in terms of scalability, cost structure, and so on. We expect that to ramp over time. Ramp is also a function of the economics of the panel in terms of yields and other things. We expect some, basically, orders to come potentially late 2027 with some revenue realization starting with 2028. But again, that would be based on the customer capacity ramp.
[Analyst 6]: Okay. Thank you. Very clear. Thank you.
Ben Poh: Thank you, Leping. Next, I would like to request Tracy. Tracy, to unmute to raise question.
[Analyst 7]: Hi. Thank you for taking my questions. My first question is about the Photonics business. Just wondering, within CPO and optical transceivers, which part of the bonding or the attachment may use more advanced equipment such as TCB, and which part may use the hybrid bonding, and why you feel it is necessary to use this advanced bonding equipment.
Bassel Haddad: You ask about the CPO, NPO?
[Analyst 7]: CPO, NPO. Yes.
Bassel Haddad: Yeah. I think, again, as we shared in Q2, we have the multiple intercept points in NPO and CPO in terms of where our machines and tools can be used. One to call out is the EIC to PIC bonding. That bonding require the accuracy and the performance of TCB, versus other bonding instances in that package. One thing with the EIC, PIC adapt technology ramps. Again, we are working with our customers to ensure that our machines are available and meet their needs, their specifications.
[Analyst 7]: Okay. Thank you. My second question is about the semi solutions gross margins outlook. Just wondering how you expect the semi solutions gross profit margin going forward.
Katie Xu: Tracy, this is Katie. Thank you for the question. I think looking ahead for semi solutions, the gross margin will continue to benefit from a relatively favorable product mix on AP. I think that we've looked at the mid-40s, right? I think, again, I take it out to the longer run, the semi gross margin could continue to expand beyond the mid-40s.
[Analyst 7]: Okay. Thank you. Thanks a lot. I do not have other questions, and will go back to the queue. Thank you.
Ben Poh: Yeah. Thank you, Tracy. Next, I would like to invite Charlotte to unmute and raise a question. Please go ahead.
[Analyst 8]: Yeah. Good morning. I have two questions. First one is on the qualification progress of hybrid bonders. Would you share any updates in terms of the qualification, and what are the applications for, whether it is C2W or W2W? I will just ask this question first.
Bassel Haddad: Yeah. We mentioned that before. Our second generation hybrid bonding solutions is competitive in alignment precision, footprint, bonding accuracy, and UPH, units per hour. We continue to work with Tier One Foundry and IDM. They are engaging with us as a second source. We believe that is due to pain points on higher precision placement accuracy and void-free bonding. We also believe that both of these customers are still very open to another player to come in. We remain very confident about the competitive capability of our second generation hybrid bonding machine.
[Analyst 8]: Yes. Got it. My second question is on the orders. Could you quantify a bit on how much within the new orders of the quarter, how much are contributed from Photonics?
Bassel Haddad: Yeah. I think, what we have seen, we mentioned that we expect the booking of Photonics to grow significantly this year, but also significantly H2 versus H1.
[Analyst 8]: Okay. Thank you.
Ben Poh: Thank you. I think we have time for one last person to raise question, and as I see from the queue, Gokul. Maybe last one or two questions.
[Analyst 1]: Yeah. Thanks, Ben, for the chance to ask a follow-up question. Maybe first one on the mainstream semi solution. The momentum seems to be pretty good, I think from your customers also. Could you talk a little bit about how you are thinking about it as you go into next year, especially given the traditional drivers of this business, like smartphone, PC, are still quite sluggish and the volumes are still declining. Are you still expecting that mainstream solutions, the order rate continues to grow into next year, solely driven by the AI power management? How do you kind of square the weakness in the consumer side of the equation?
Bassel Haddad: Oh, thanks, Gokul. Yeah. We have seen demand in our mainstream solutions from AI data center to what you mentioned, mostly around power, with IDMs continuing to invest in their products to really cater to the AI demand. By the way, that also drives business for SMT as well in terms of the board side of things for the power for AI. We were monitoring the situation on the end market and consumer. So far we have not seen order cuts from consumer, but we continue to monitor that. Basically, our booking on mainstream remains elevated, even in Q4, supported by the AI infrastructure expansion and also China, OSAT experiencing high utilization. I want to call out also that our exposure is much broader than consumer, right? We are not only solely tied to consumer.
Bassel Haddad: We have a diverse customer base across automotive, industrial, power, communication, and computing end markets.
[Analyst 1]: Got it. Understood. Just to follow up on your second source commentary for hybrid bonding, for the leading IDM and Tier One Foundry. Could you give us a little bit more color on this? Is this a recent development that you become the potential second-source vendor here? Any indications you are getting from the customers for potential for high volume shipments, given that especially for the Tier One Foundry, the big ramp is going to be H2 next year and 2028.
Bassel Haddad: Yeah. I want to tie it to what I said. We are going to be technology-led, and we will look for where we can differentiate our products and be there for our customer, and win market share, even if they have an existing leading supplier. We are working with them. We believe every big customer in such a new technology would look for more than one supplier. We are counting on being the second supplier that cater to that hybrid bonding demand growth that would happen probably in the next 12 to 18 months. We start seeing some movement into qualification, Gokul.
[Analyst 1]: Got it. Yeah, that is very clear. Thank you very much.
Ben Poh: Thank you, Gokul. I think that will be all for the questions for today. This will conclude our call. Thank you very much for participating and take care. See you guys again. Bye-bye.
Operator: Goodbye.
