慧榮科技 (SIMO) 2026 年第三季法說會：企業級業務成長推升營收創新高

重點提要

• 慧榮科技 (NASDAQ: SIMO) 公布 2026 年第三季營收達 5.6 億美元，季增 24%、年增 131%。營收超出公司原本預期的 5.19 億至 5.41 億美元指引區間。
• Non-GAAP 毛利率達 53.7%，營業利益率升至 32.1%。每單位 ADS 盈餘季增 79%、年增 333% 至 4.35 美元。
• Ferri 車用產品、企業級開機碟，以及 Edge 和 MonTitan SSD 控制晶片帶動成長。儲存解決方案佔當季營收的 40% 以上。
• 管理層預測 2026 年第四季營收為 5.88 億至 6.16 億美元，代表季增 5% 至 10%。毛利率預計為 54% 至 55%，營業利益率為 36% 至 37%。
• 展望 2027 年，在 MonTitan、Ferri 車用產品和企業級開機碟的支撐下，管理層預期營收將成長至少 40%。毛利率預計落在 50% 出頭至 55% 的區間。
• NAND 與 DRAM 價格高企仍是智慧型手機、PC 及其他消費性裝置的逆風，但管理層預期企業級與車用業務的成長將降低公司受消費性季節性影響的程度。

核心財務數據

慧榮科技於 8 月透過可轉換公司債籌集了 11.5 億美元，淨收益為 11.27 億美元。該公司債票面利率為 0%，初始轉換價格為 380.50 美元。管理層表示，隨著儲存解決方案業務規模擴大，發行所得將可充實營運資金容量。

業務與營運表現

Ferri 車用與企業級開機碟

Ferri 及開機碟解決方案在第三季表現最強勁，營收季增超過 75%，佔總營收的 40% 以上。在長產品週期和持續擴大的設計獲案 (design wins) 支撐下，車用產品帶來優於預期的營收與更高的毛利率。

管理層表示，公司正與多家機器人公司合作。Ferri 的車用級能力（包含耐高溫與抗震）也能支援人形機器人、無人機及其他實體 AI 系統。然而，管理層強調，機器人業務仍處於早期階段，目前出貨量有限。

慧榮科技預估，2026 年無 DRAM (DRAMless) 企業級開機碟市場規模約為 30 億美元，到 2030 年將成長至 50 億美元以上。公司亦看好搭載板載 DRAM 的近 CPU (near-CPU) 開機碟新興市場，管理層預估其規模到 2030 年可能突破 100 億美元。本身的近 CPU 解決方案預計將於 2027 年下半年開始放量。

MonTitan 企業級 SSD 控制晶片

MonTitan 繼續在 NAND 品牌廠、超大型雲端業者、雲端服務提供者及模組廠中獲得採用動能。前兩家客戶已於 2026 年第二季開始出貨使用 MonTitan 的 TLC 企業級 SSD，管理層表示已公布的全部七家客戶均按計畫於 2026 年期間進行商業化放量。

慧榮科技在第三季完成下一代 4 奈米 PCIe 6 MonTitan 控制晶片的 Tape-out（光罩下單/試產）。預計於 2027 年第一季開始送樣。該控制晶片鎖定 TLC 運算應用與 QLC 資料儲存應用。

管理層預期，由於 2Tb QLC 晶粒供應有限，短期內 TLC 運算 SSD 的成長速度將快於 QLC 儲存 SSD。六家 NAND 大廠中有四家已開始生產 2Tb QLC 晶粒，公司預計所有 NAND 大廠都將在 2027 年底前出貨此類產品。

Edge SSD 控制晶片

在四通道與八通道 PCIe 5 產品的支撐下，SSD 控制晶片業務展現強勁的季增與年增成長。根據管理層數據，PCIe 5 約佔 PC 市場的 10%，預計到 2027 年底滲透率將達到 20% 至 25%。

然而，對於第四季，由於記憶體和儲存市場狀況以及正常季節性因素，公司預計 SSD 控制晶片營收將持平或略微下降。

eMMC 與 UFS 控制晶片

第三季 eMMC 與 UFS 營收較上一季小幅下滑。管理層將此下降歸因於客戶新品推出時程，以及 NAND 與 DRAM 價格上漲造成的負擔能力壓力。

公司預計該業務短期內將保持相對持平。管理層認為隨著 UFS 4.1 和 UFS 5 控制晶片放量出貨並支援更高的平均售價，業務有機會恢復成長。

管理層指引

管理層預計 2026 年營收將較去年同期翻倍以上，創下公司歷史新高。管理層亦預計今年結束時營業利益率將超過 30%。

展望 2027 年，成長預計主要來自 MonTitan 企業級控制晶片、Ferri 車用產品、開機碟及其他 SSD 控制晶片。管理層表示，明年儲存解決方案佔營收比重可能接近 50%，在解決方案與控制晶片之間形成更平衡的組合。

風險與關注焦點

• NAND 與 DRAM 價格高企正在給 PC 和智慧型手機的負擔能力帶來壓力，導致 OEM 廠商降低裝置規格，並導致消費性產品出貨量下降。
• 管理層預計記憶體、晶圓、基板、封裝與測試產能仍將受到限制。公司表示已向台積電確保了 2027 年主要的晶圓需求，以及測試、封裝和基板產能。
• 儲存解決方案比重提升增加了對 NAND 採購、庫存及價格的曝險。慧榮科技採用六個月期的客戶定價安排，並以從採購 NAND 到產生現金流約需要兩個季度為目標。
• 近 CPU 開機碟需要客製化韌體與客戶特定功能。管理層表示，研發資源有限可能會影響公司拓展新機會的速度。
• 在 3 月進行結辯後，新加坡仲裁案仍未解決。管理層並未提供作出裁決的時間表。

分析師問答重點

2027 年成長與毛利率：管理層將營收成長至少 40% 的預測歸因於 MonTitan 部署的擴大、Ferri 車用產品強勁放量以及新的開機碟專案。多年期車用認證與更豐富的企業級產品組合預計將支撐毛利率落在 50% 出頭至 55% 的區間。

供應鏈產能：慧榮科技表示已主要向台積電確保 2027 年的晶圓供應，以及所需的測試、封裝和基板產能。管理層相信這能夠支撐其目前的成長計畫。

NAND 價格曝險：公司透過協商好的定價安排將 NAND 成本上漲轉嫁給客戶。管理層亦認為，在近期規格調降後，若 NAND 價格下降，可促使客戶恢復儲存容量密度。

營運資金：根據當前預測，可轉換公司債發行所得預計將支應未來幾年的營運資金需求。第四季庫存預計將繼續增加，隨後隨著產品出貨轉化為現金。

近 CPU 開機碟：慧榮科技正在與模組廠、NAND 供應商及 Tier 1 伺服器客戶合作。透過合作夥伴出貨的控制晶片最早可能在 2027 年第二季開始，而公司的完整解決方案預計將在 2027 年下半年放量。

中國策略：管理層表示，慧榮科技計劃在擴大在地研發的同時，支援中國 NAND 廠商長江存儲 (YMTC)、模組廠、超大型雲端業者與雲端服務提供者。公司打算供應企業級控制晶片與韌體，而非製造完整的企業級 SSD 解決方案。

法說會逐字稿全文

---

完整財報電話會議逐字稿

管理層陳述

Operator

Good day, and thank you for standing by. Welcome to Silicon Motion Technology Corporation's Third Quarter 2026 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. This conference call contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933 and Section 21E of the Securities Exchange Act of 1934 as amended.

Such forward-looking statements include, without limitation, statements regarding trends in the semiconductor industry and our future results of operations, financial condition and business prospects. Although such statements are based on our own information and information from other sources we believe to be reliable, you should not place undue reliance on them. These statements involve risks and uncertainties, and actual market trends and our results may differ materially from those expressed or implied in these forward-looking statements for a variety of reasons.

Potential risks and uncertainties include, but are not limited to, continued competitive pressure in the semiconductor industry and the effect of such pressure on prices, unpredictable changes in technology and consumer demand for multimedia consumer electronics, the state of and any change in our relationship with our major customers and changes in political, economic, legal and social conditions in Taiwan.

For additional discussion of these risks and uncertainties and other factors, please see the documents we file from time to time with the Securities and Exchange Commission. We assume no obligation to update any forward-looking statements, which apply only as of the date of this conference call. And with that, I'll now hand you over to Mr. Tom Sepenzis, Vice President of Investor Relations and Strategy. Please go ahead.

Thomas Andrew Sepenzis

Good morning, everyone, and welcome to Silicon Motion's third quarter 2026 financial results conference call and webcast. Joining me today is Wallace Kou, our President and CEO, and Jason Tsai, our CFO. Wallace will first provide a review of our key business developments, and then Jason will discuss our third quarter results and outlook. Following our prepared remarks, we will conclude with a Q&A session.

Before we begin, I would like to remind you of our safe harbor policy, which was read at the start of this call. For a comprehensive overview of the risks involved in investing in our securities, please refer to our filings with the U.S. Securities and Exchange Commission. For more details on our financial results, please refer to our press release, which was filed on Form 6-K after the close of the market yesterday.

This webcast will be available for replay in the investor relations section of our website for a limited time. To enhance investors' understanding of our ongoing economic performance, we will discuss non-GAAP information during this call. We use non-GAAP financial measures internally to evaluate and manage our operations. We have therefore chosen to provide this information to enable you to perform comparisons of our operating results in a manner consistent with how we analyze our own operating results. The reconciliation of the GAAP to non-GAAP financial data can be found in our earnings release issued yesterday. We ask that you review it in conjunction with this call. With that, I will turn the call over to Wallace.

Chia-Chang Kou

Thank you, Tom. Hello. Thank you for joining us to the call today. I am pleased to report that we delivered another quarter of strong growth with revenue of $560 million and gross margin of 53.7%, driven by the growth in Ferri for automotive, Edge AI, and boot drive storage, our Edge SSD business, and our emerging MonTitan controller for enterprise SSD. We drove another quarter of higher operation performance, delivering the second consecutive quarter of record earnings per ADS, as we continue to capitalize on our product and market diversification strategy. We are on track to deliver record revenue in 2026 with over 100% top-line growth. Our strategy is simple, deliver real value to customers and focus where we are uniquely differentiated so we can generate strong ROI for shareholders. For our edge controllers, we have become the de facto standard for mobile, IoT, and PC controller solutions.

For NAND makers and virtually every module maker with outsourcing become a much more meaningful portion of the market. Our matched controller technology and design expertise deliver solutions that are better performing, consume less power, and are more cost-efficient than either in-house or other merchant solutions. Our strong share gains are driving significant growth despite end market weakness, and we expect this to continue.

Our MonTitan enterprise SSD controllers are winning across the ecosystem with leading NAND makers, hyperscalers, and CSPs, and with the module makers that supply AI server to hyperscale and enterprise customers. Our success is based on our unique combination of two decades of controller and firmware design experience and our history of managing more NAND than anyone in the industry.

This has created an incredible, solid foundation for our leading enterprise controller products. And then we expect our MonTitan revenue ramp to accelerate as we introduce our next generation PCIe 6 controller next year. And lastly, for our storage solution segment, our strategy has always been to focus in areas that are underserved by NAND makers. We are rapidly growing share in this vital low density, smaller volume market that require more design and customization work. We have spent more than a decade building one of the industry's broadest automotive-grade portfolio, supported by dedicated quality and supporting team. Today, that foundation positions us to capture fast-growing compute and storage demand across cars, drones, and increasingly robotics.

For our enterprise boot drive solution, our unique combination of broad controller platforms, best-in-class firmware, technical know-how, NAND suppliers relationship, and manufacturing capability put us in an ideal position to serve the growing need of AI infrastructure market. As a result, we are rapidly evolving to much more balanced and profitable company as a new enterprise in AI product enhance our leading consumer controller share, strengthening our competitive position from AI infrastructure to the edge to deliver long-term sustainable revenue and earning growth. I would like to take a moment to address the current market environment. The AI super cycle has driven robust demand of HBM, DRAM, NAND, HDDs for the past several quarters, with no sign of slowing down. Prices have risen sharply over the past year and the strained substrate supply across memory and storage enterprise market remain relatively insensitive to price.

For PC, smartphone, and other consumer device OEM are struggling to bring affordable product to market, given today's memory and storage cost, and the consumer unit volume are declining as a result. We expect conditions to stay difficult until bit supply come back into balance with demand. New NAND fabs are planned for 2028, and NAND makers continue to add bit through higher layer count. So we do not see this changing the current supply-demand imbalance to any great degree. Flash makers continue to prioritize enterprise and high-capacity solution, creating new and expanding opportunity for Silicon Motion. Our growth is benefiting from more outsourcing and larger opportunity in enterprise and the storage solution market that we are uniquely positioned to win. We are at the beginning of the journey in the enterprise and AI market, and our early success will drive accelerating demand for our next generation storage technologies.

Let me now detail our embedded eMMC and UFS business, which include controller for smartphone IoT connected devices. Our eMMC, UFS business was up over 100% year-over-year in the first half of 2026 as the NAND makers are shifting their resources toward DRAM and HBM solutions, allowing module makers to take share in the mobile market. This has helped drive our market share gains as module makers look to third-party controller partner to make their mobile solution.

In the third quarter, eMMC and UFS revenue declines modestly from the prior quarter, reflecting customer launch timing and the challenging smartphone OEM face in offering affordable device amid the higher NAND and DRAM prices. We expect the headwind to persist in the near to midterms, until additionally memory and storage supply comes online.

Across the markets that we sell our eMMC, UFS controller, OEM are reducing specification to try to offset the rising memory and storage cost. We expect our eMMC, UFS business to remain relatively flat in the near term given the current market challenging, but I do believe we can return to growth next year and beyond as our new UFS 4.1 controller and UFS 5 controller ramp and driving higher ASPs.

Now turning to our SSD controller business, which include Edge and MonTitan enterprise SSD controllers. This business delivers strong sequential and year-over-year growth. In SSD, our PCIe 5 investment continue to pay off, with growth in both our four-channel and eight-channel controllers. PCIe 5 currently represent approximately 10% of the PC market, and we expect this will grow to 20%-25% of market by the end of 2027.

Providing a strong runway for further growth. Our leading position in PCIe 5 is driving greater market share gains within the PC market. And as the industry transition from PCIe 4 to PCIe 5, we expect to see further top-line improvement given the higher ASP associated with these next-generation products. We are currently working with all the major NAND makers and most of the leading module makers to deliver compelling SSD solutions that leverage PCIe 4 and PCIe 5 controllers. We are tracking well ahead of expectation to achieve our market share goals of 40% through Apple by 2028, and believe we will continue to improve our position. While PC unit headwinds associated with the higher memory and storage prices present challenges, we believe we will drive further top-line growth through the market share gains and ASP increases as the market shift to PCIe 5 continues.

I would now like to provide you with an update on our new MonTitan enterprise SSD business. NAND remains a critically important technology across enterprise and AI storage ecosystem, and MonTitan is positioned for rapid growth in the coming years. This eSSD solution delivers high speed, low latency storage that targets compute and data storage application for enterprise AI platforms.

Our first two customers began shipping TLC eSSD for compute that leverage our MonTitan controller in the second quarter. All seven of our announced customers are on track to begin commercial ramp in 2026 with both TLC-based compute and QLC-based data storage products. TLC-based compute eSSD using MonTitan are expected to grow more quickly than QLC-based data storage eSSD in the near term, until greater supply of 2 terabit die become available.

Encouragingly, four of our six NAND makers have started production of 2 terabit QLC die to date, and we expect all NAND makers to begin shipping 2 terabit QLC by end of 2027. So momentum will accelerate through next year. During the third quarter, we taped out our next-generation 4nm PCIe 6 MonTitan controller targeting hyperscalers and CSPs, and expect to begin sampling in the first quarter of 2027. We have one design with multiple tier one customers, including NAND flash makers, hyperscalers, and CSPs. The new controller was designed for both compute application using TLC and data storage using QLC NAND as a performance shaping and multi-tenancy service architecture extremely well suited for a Gen AI infrastructure. We continue to see increased inbound interest in our MonTitan controllers, and expect this new enterprise controller segment will help us deliver long-term sustainable top and bottom-line growth.

And finally, I would like to provide an update on our Ferri automotive, edge, and enterprise boot drive storage business. Our solution business, which includes Ferri and boot drives, delivered the strongest result in the third quarter, with revenue up over 75% sequentially, representing over 40% of our revenue. Our automotive business has been a particularly strong source of strength, delivering better-than-expected top-line results and higher gross margins. We continue to scale this business quickly as we know automotive market, its customers, and the supply chain extremely well, which has driven significant backlog for long-term growth as we accelerate our share gains. We have since spent more than a decade developing our automotive products and certifications. Today, our storage controller and firmware offer the broadest portfolio of solution to serve every part of the automotive market.

Our success in automotive has drawn interest in Ferri for other large growing markets, including robotics, drones, and advanced networking. In the emerging robotic market, we are actively engaged with several companies that want to use our storage product. We are seeing multiple opportunities in humanoid robots and other physical AI systems, including vision, LiDAR, and compute storage.

Our early conversation suggests robotic opportunity may be larger than automotive, and we believe Ferri is ideally suited to address it. Our fast-growing boot drive business is also gaining momentum. We estimate that the market will be approximately $3 billion in 2026, growing to over $5 billion by 2030. In the third quarter, we won design with several new Tier one customers for networking and AI infrastructure products. These programs are expected to ramp over the next 2 years and will substantially improve our backlog visibility.

As the average capacity of enterprise boot drives is much smaller than compute and data storage SSD drives, the NAND makers no longer prioritize this market. Additionally, our DRAMless controller delivers the performance, reliability, endurance this drive needs with lower power consumption requirements, smaller form factor, and the lower cost than DRAM-based design. Boot drive handles the operation system, logging, and system management rather than heavy data workload, so they do not need the expense of onboard DRAM for DPU, router, and switch products. As AI infrastructure builds out at large numbers of servers, the networking and DPU equipment shipments are increasing rapidly, each needing reliable boot storage, driving the growing demand for DRAMless boot drives. We have been the primary supplier for DRAMless boot drive controllers to the NAND maker that have historically served this market, but they are moving on the higher density opportunity.

Silicon Motion are uniquely positioned with the right technology, the right support, the leading security firmware, and proven ability to procure NAND, making us the best partner for those needing a boot drive solution. Looking ahead, however, we are seeing significant new opportunity for boot drive with onboard DRAM. Given the strong demand of AI inference and agentic AI, near CPU boot drive with DRAM provide lower latency and open compute project features that are in increasingly high demand. As the NAND maker have shifted resources toward HBM, DRAM, and higher capacity compute and data storage SSD, there is a growing shortage of near CPU boot drives with DRAM. We estimate that this market will reach over $10 billion by 2030, making a significant new opportunity for Silicon Motion.

We are already in discussion with multiple potential customers for boot drive that leverage our industry-leading eight-channel and PCIe 5 controller that can serve the near CPU market to further expand our market opportunity, and expect to begin ramping product for this market in the second half of 2027.

Our Ferri automotive, edge AI, and boot drive solution business are performing especially exceptional well, and these are many large and growing opportunity for our product in this segment.

Our backlog has increased significantly in just the past three months, and we expect many additional customer design wins in automotive, edge AI, and boot drive solution to deliver significant future growth. We continue to benefit from our proven ability to secure NAND in a difficult environment from multiple NAND suppliers, and this continues to be a powerful differentiator for our company.

In conclusion, Silicon Motion delivered its fourth consecutive quarter of record revenue as we executed across a rapidly diversifying business. We are a fundamentally stronger company today with a stronger balance sheet, broadest range of solution and product, and more opportunity ahead than at any time in our history. Our broad portfolio serves surging AI demand from the data center to the edge. That breadth gives us balance across our market and the flexibility to seize growing opportunity while navigating challenge like those we face today in PC and smartphones. I am extremely proud of our teams. Together, we have built a durable, diversified business powered by best-in-class technology, expanding share and new end markets, strengthened by two decades of relationship with OEMs, module makers, and NAND makers. I am more confident than ever that we will continue to deliver broad-based, sustainable growth long term. Now let me turn the call to Jason to go over our financial performance and outlook.

Jason Tsai

Thank you, Wallace, and good morning to everyone joining us today. I will discuss additional details of our third quarter results and then provide our outlook. Please note that my comments today will focus primarily on our non-GAAP results, unless otherwise specifically noted. A reconciliation of GAAP to non-GAAP data is included with our earnings release issued yesterday. Before I get into the results, I would like to comment briefly on our convertible bond issuance on August 13th. We raised an aggregate $1.15 billion with net proceeds to the company of $1.127 billion. This successful raise was increased from the initial offering of $800 million. This new CB at 0% interest and an initial conversion price of $380.50 represents a conversion premium of 65% over the last reported sale price of SIMO on August 10th, the day the deal was announced.

This additional capital significantly strengthens our balance sheet and will help fuel our growth and enhance our working capital to scale our business. Now, turning to our results. We delivered an outstanding third quarter, surpassing our expectations, driven by stronger execution and demand trends. In the September quarter, sales increased 24% sequentially and 131% year-on-year to $560 million, coming in well above the high end of our guided range of $519 million to $541 million, delivering our fourth consecutive quarter of record revenue. We experienced strong growth across most of our businesses in the third quarter, with standout growth in Ferri for automotive, enterprise boot drives, and SSD controllers, including Edge and MonTitan for enterprise. Gross margins of 53.7% was significantly better than guidance as we capitalize on new product introductions and benefited from better mix and better-than-expected margins in our Ferri automotive products.

Operating expenses slightly decreased sequentially to $120.7 million and operating margin increased 32.1%, well exceeding our guidance, driven by higher-than-expected revenue and gross margin in the quarter. Earnings per ADS was $4.35, a 79% increase sequentially and a 333% increase compared to same quarter last year. Total stock-based compensation, which we exclude from non-GAAP results, was $17.3 million in the September quarter. We had $953.6 million in cash equivalents, and restricted cash at the end of the third quarter, compared to $181.8 million at the end of the second quarter. Cash increased in the third quarter due to the successful $1.15 billion convertible notes offering that we executed in the quarter, offset by the increase in working capital, dividend payment, and the repayment of the short-term bank loan.

Our strategy, as Wallace outlined at the start of this call, continues to drive growth and expand our opportunities. Our experience, our customers and partners, and our unmatched technology and development of NAND controllers and solutions has driven a significant transformation of the company. While the NAND market continues to be dynamic, our investments across consumer, enterprise, data center, and AI applications are the foundation of our rapidly growing business.

As a result, we believe the momentum in our business will continue in the fourth quarter and now expect revenue to grow 5% to10% sequentially to $588 million to $616 million, with growth across nearly all our product segments led by Ferri for automotive, enterprise boot drive solutions, and MonTitan enterprise SSD controllers. We believe our SSD controller business will be flat to slightly down given the current memory and storage environment and typical seasonality. Gross margins are expected to increase sequentially to 54% to 55% in the December quarter, given the continuing improvement in product mix.

Operating margin is expected to grow to 36%-37% as continued strong revenue growth and higher profitability drives the leverage to the bottom line. Our effective tax rate is expected to be 20%, and stock-based comp and dispute-related expenses are expected to be in the range of $34.6 million to $35.6 million. 2026 is on track to deliver record revenue for Silicon Motion, with top line expected to more than double year-over-year. While we continue to invest heavily in R&D this year to expand our portfolio of leading-edge solutions, we are confident that along with the higher revenue and improved gross profitability, our operating margins will exceed 30% exiting this year. We expect this momentum to carry into 2027.

Our strong backlog across all of our businesses gives us confidence that we can grow revenue by at least 40% next year, with gross profitability to be in the low to mid 50% range. Our mix is improving. With more contribution from enterprise controllers and Ferri automotive solutions, we should be able to sustain this higher margin level. It is also worth emphasizing that these new enterprise opportunities represent larger addressable markets than our consumer segments.

And because they are greenfield opportunities for Silicon Motion, we can grow significantly by gaining share without depending on further market growth. Any market expansion would only add to our potential upside. We set out to diversify beyond consumer applications into automotive and enterprise, and that strategy is now producing outsized results. Our longstanding leadership in merchant controllers and our multiyear investments in enterprise and AI SSDs are beginning to bear fruit.

The solutions we are winning in these markets offer far greater visibility, much longer product cycles, and significantly higher barriers to entry, which together underpin sustained growth in revenue and profitability. As these wins ramp across a broad mix of end markets, consumer cycles will have less influence on our results. Our visibility is improving significantly, and our business will become meaningfully more resilient and predictable. Our balance sheet has never been stronger, and we are well-positioned to capitalize on the fundamental drivers at our foundation of the transformation that Wallace spoke about earlier. I look forward to sharing more on our progress next quarter. This concludes our prepared comments. I would like to open up the questions now. Operator?

Operator

[Operator Instructions] Our first question comes from the line of Neil Young of Needham & Company.

分析師問答

Neil Young

First question, just given that you just gave us the expecting revenue to grow by at least 40% next year and gross margin in the low 50% range, curious if you could sort of dive into maybe the main drivers of that. I mean, the given that gross margin, you expect sort of resetting the long-term target, it sounds like now. I would expect more of the growth to be driven by sort of the higher margin business like Ferri and MonTitan, but any comments that you can add further on to '27?

Jason Tsai

Yes, I think that is right. I think we are seeing significant traction with MonTitan. We are seeing the automotive, Ferri for automotive, ramping very aggressively as well. A lot of these Ferri projects, especially in automotive, these are multi-year engagements where, once we are qualified in, we have got a really long-standing position with that socket, with that customer. We are working with virtually every automaker in the world for a variety of solutions going into a significant amount of their portfolio, but significant room for growth still. So based upon our backlog, based upon the growth we anticipate from MonTitan, as well as other SSD controllers and Ferri for automotive, we believe that we can sustain these low to mid 50% gross margins.

Neil Young

Okay, great. That is helpful. Just in this quarter, bringing it back, looking at the Ferri and boot drive, obviously the growth continuing is really great. Could you, by any chance, give us some more detail? It sounds like Ferri might have outgrew boot drive in the quarter. Obviously, beyond automotive, I should say, in Ferri, curious to learn more about the robotics and where you guys play in that.

Chia-Chang Kou

I think because of Ferri for automotive been for a long time, so we see the custom base are very strong. For this year, definitely it is Ferri for automotive, it grows stronger. However, our boot drive also gaining momentum with the diversified customers, more than just GPU customer and our TPU customer. We recently winning all the networking and server customers. We see our enterprise boot drive will grow even much faster next year.

In robotics, I think it is still very early stage, and the total volume is still very small, but we try to make sure we will be in when the market take a boom.

Jason Tsai

And keep in mind, a lot of the features within the Ferri for automotive products are the same features that the robotics requires, right? Which is high temp, vibration, all of the things that go into a car largely go into the robotics as well. So having that certification already, having those products already, puts us in a significant advantage to penetrate this market as it grows.

Operator

We will now take our next question from the line of Craig Ellis of B. Riley Securities.

Craig Ellis

Congratulations on the tremendous momentum that it seems you have in the business. I wanted to start with a question for you, Wallace. Very exciting to hear about the engagement that you have got with near CPU boot drives, because the performance gains, given reduced latency with the way that can implement are really quite profound, and it seems like that could be a very significant market. The question is this: Do you see this implementing on more of a custom basis across customers, or is JEDEC going to move super quickly and get standards out here in another quarter or two? Then, can you just talk a little more about how you would see the early part of the ramp for Silicon Motion?

Chia-Chang Kou

Now we see opportunity come from the early third quarter, because conventionally, this near CPU boot drive with DRAM belong to NAND makers. Because all NAND makers are busy and try to produce at least eight terabyte, 16 terabyte Compute SSD. They have less interest of resource on NAND allocation for the low capacity near CPU boot drive. The near CPU boot drive primary density is about one terabyte, two terabyte, and some also 512 gigabyte. four terabytes, very, very few. I think because we can see it will be probably a very, very big gap of shortage supply for 2027. Many, many customers come to Silicon Motion. We are preparing discussion, and I think the solution, definitely need to have a customized sum feature and tailored for individual server and leading hyperscalers. So this is not a simple work.

We will try to focus on select the NAND and make sure we follow the OCP requirement and provide the firmware. I think we will be ready for the firmware and solution by early Q1, and production ramp will be second half 2027.

Craig Ellis

That is really helpful. The second question is really a higher level question for you and Jason. The team has done a phenomenal job operationally this year, putting the business in a position to grow 100% plus year-on-year. Can you help us understand the positioning that you feel like you are in with wafer supply, packaging, et cetera, to support the 40% growth? If demand is greater, given some reports that things like agentic AI, CPUs could grow 47% next year, can you just help us understand if there is upside potential in the supply chain to meet greater demand than that 40%?

Chia-Chang Kou

As you know very well, in semiconductor industry today, everything is in shortage, right? From substrate, from material, from wafer, from NAND, from memory, from even test capacity. But I think our manufacturing operating team did a very good job. We secured our 2027 wafer with TSMC primarily. And also, we also secured the test capacity and assembly capacity and the substrate material. So we believe we are able to support our growth for 2027. Jason mentioned we see we definitely can grow minimum 40% year-over-year. I think in next quarter earning call, we will provide more meaningful guidance for our 2027.

Operator

We will now take our next question from the line of Sebastien Naji of William Blair.

Sebastien Cyrus Naji

Congrats on another record quarter here. My first question is maybe a little bit higher level and strategic. As your SSD solution business continues to grow as a share of revenue hit over 40% in Q3, it is driving greater NAND pricing exposure into Silicon Motion's model. And that is a little bit of a shift for a company that was largely tied to volume growth historically. So maybe just how do you think about managing this new pricing dynamic and how it could impact your growth and margins, particularly if NAND pricing starts to come down?

Chia-Chang Kou

I think we definitely manage the NAND price increase and negotiate with the customer to pass through the cost to customers, so we benefit from the sale revenue growth. So we believe the NAND price increase will come to moderate because the spot price for module maker now has become challenging. We see the Q3, Q4 NAND pricing are rather similar. For next year, we also see the, although will be a single digit growth quarter, but it will be moderate. For our business, why we focus on automotive and boot drive, because the end customer want to have a long-term contract and support and supply. Our contract with customers and price will stay with at least two quarter, and we negotiate price every 6 months.

We are able to manage the price and the cost, and that is why we can see through the demand, and we can manage our gross margin continually. We see the NAND wafer price will be more stable in comparison last year.

Jason Tsai

One more thing, Sebastien. I think certainly there is a fear that, if NAND flash prices begin to come down, how does that affect our growth? I think one of the things that we are seeing in automotive, even in boot drives, is that there is some just despeccing going on. I think all things being equal, if NAND availability was better, if pricing was better, our customers would double their density or more in some cases. I think between a combination of kind of respecing to normal specs as well as additional opportunities, additional customers that is coming online over the next 12 months, 18 months, 24 months, I think we are well positioned to continue to grow even as the NAND market stabilizes over the next 2 years.

Sebastien Cyrus Naji

Got it. Great. That's helpful. Maybe just as a related follow-up around the funding that you raised and the working capital needs, I just wanted to get maybe a better sense for your ability to continue funding all of your working capital and inventory build with the cash that you have on hand. Can you maybe just talk a little bit about how quickly you expect inventory, particularly around NAND, to flow through the model and what it would take to maybe potentially need further capital down the line?

Jason Tsai

Yes. For us, we're targeting about two quarters turnaround time between when we get the NAND and when we generate the cash flow. So, our expectation is to be able to turn that inventory relatively quickly, and generate the cash flow to support ongoing growth of our business. But when we did the raise, we talked about the raise being sufficient to fund our working capital needs for the next couple of years, based upon today's expectations. Obviously, as the business continues to grow and hopefully as it exceeds expectations, we'll continue to reevaluate that. But that's, at least right now for our visibility, we believe we're in a good spot.

Operator

Our next question comes from the line of Suji Desilva of ROTH Capital.

Sujeeva De Silva

Congratulations on the progress here. The MonTitan product, the higher KV cache density you're expecting, are there innovations coming from HBM and DRAM that might offset the opportunity for SSDs and KV cache? I'd love to get your thoughts there. What's a good concept of the number of MonTitan and SSDs needed per GPU for KV cache?

Chia-Chang Kou

I think the MonTitan demands from very wide range, not only from the AI server and for hyperscaler and also for the CSP, but also for NVIDIA ecosystem. I think we don't know the percentage how, but we see the demand for Compute SSD this year is bigger due to the NAND price is very expensive. But we also see the momentum for high capacity QLC coming for next year. I believe, when Samsung, Hynix, 2TB QLC can move to production, we'll see more QLC demand for data storage. MonTitan itself can handle both the Compute SSD and the Storage SSD. For KV cache, particularly for CMX, for NVIDIA structure, I think they will redefine and try, because the price is very expensive, they try to reduce the minimum capacity for each of their route and trade.

I think this will be probably, they will announce it early next year, and I think the related ecosystem will help him provide. I think as well as the CMX KV cache demand is growing, as well as demand additional eSSD. That's good for us, for MonTitan, because it increase our controller demand.

Sujeeva De Silva

Okay. Very helpful. The other question's for Jason, perhaps. 40% plus revenue growth next year, you talked about an operating margin guide in the 30s. Just how are you managing the OpEx and R&D growth, given that you have so many different end markets and projects you're focusing on? What's a kind of a growth rate for OpEx versus revenue that would be kind of a framework for us to think about?

Jason Tsai

Yes. We'll get into more specific guidance around 2027 in January. This year we taped out a number of 6 nano meters and as well as 4 nano products. So, we're spending a lot of money on tape-out and design this year already. We anticipate continuing to invest next year. And again, we'll talk more about the specific OpEx guidance next year in January, but our expectation is that we will continue to be able to deliver operating margin leverage next year as well.

Operator

Our next question comes from the line of Mehdi Hosseini of Susquehanna International Group.

Mehdi Hosseini

Yes. A couple of follow-ups from me. Wallace, Jason, can you quantify what percentage of your total revenue were attributable to Ferri, boot drive, and PCIe SSD, and how each percentage would change into next year? I have a couple of follow-ups.

Chia-Chang Kou

So I would say we cannot party about detail, but about 40% our total revenue is around the solution from Boot Drive and Automotive, 60% from controller.

Mehdi Hosseini

Would those mixes change dramatically next year?

Chia-Chang Kou

Next year, the solution Boot Drive Automotive might grow to 50%, but so far we didn't accurate, but I think that we see probably will be balanced next year, 50/50 between controller and solutions.

Mehdi Hosseini

Okay. Given how your exposure to the consumer end market is less looking forward compared to prior years, should I assume that there is no more seasonality from Q4 to Q1?

Jason Tsai

Again, Mehdi, we will get into more specific Q1 and 2027 guidance in January.

Mehdi Hosseini

Okay, I was hoping to get a preview. You mentioned you are more driven by enterprise, and I am just trying to, maybe qualitatively, you can help us to.

Jason Tsai

No, I can. Certainly, I think the enterprise becoming a bigger portion of the mix, Ferri becoming a bigger portion of the mix, I think certainly normal consumer seasonality will certainly be more muted going forward as consumer business becomes a smaller percentage of our business.

Mehdi Hosseini

Okay. And then is there any update on arbitration? It would be great to tell us whether the key milestone and how we track into those.

Jason Tsai

Yes. We did the closing arguments in Singapore arbitration in March, and we are still anxiously awaiting the results. We don't have any other updates at this point, but certainly once the results come out, we'll be talking about those things quite a bit. But right now, there's no additional comments we're making around the arbitration.

Operator

[Operator Instructions] Our next question comes from the line of Gokul Hariharan of JPMorgan.

Gokul Hariharan

The first question is on the near CPU boot drive opportunity. Wallace, I think you said you will start your solutions in the second half of next year. Could you talk a little bit about how big that opportunity can scale into? Because your enterprise boot drive partnership with NVIDIA, it scaled up very rapidly. But could you talk a little bit about how the opportunity is and how fresh that opportunity is, given that there are already a fair number of incumbents there? I just wanted to understand, the TAM is big, but how quickly can you scale this business?

Chia-Chang Kou

I think that 10, will be a minimum $10 billion by 2030, because the density is bigger than the DRAMless boot drive. Because the NAND maker did not pay attention, but this business belonged to a NAND maker in the past, because especially the three NAND maker have DRAM. But because all the NAND makers, they focus on high capacity compute SSD, and the data storage SSD. So there is lack of resource and attention in the near CPU boot drive with DRAM. So that's why the supply becomes shortage, and many tier one customer come to Silicon Motion. We see the great opportunity. We also work with the leading NAND maker to see how we can work together to resolve the issue. So whether they will support more NAND. And, also we need to resolve manufacturing and also including some DRAM supply, too.

But we believe we'll start to ramp in the second half of this year. We also work with some module maker partner, because they will use our controller in the firmware, but we will have a different business model to work with them to benefit some financial results. Anyway, this is a new opportunity, and we just really don't have enough R&D resources. There's a great opportunity for Silicon Motion to grow, and they could grow even faster in '28 and '29. Because we have the great idea controller and fitting very well, and we have the firmware technology, and suddenly the controller, very lucky, have a certain OCP feature built in, so we don't need to redo the new ASIC. I think a lot of our NAND makers, they don't have a such controller today. So that's why we are in a great position to expand to the new market.

Gokul Hariharan

Got it. Just to follow up there. When you win business, do you have to win by individual server customer, or you just design into a chip platform and you kind of become the de facto in that chip platform, similar to what you do for the current enterprise boot drive?

Chia-Chang Kou

All the near CPU boot drive today is M.2 form factor, 22 by 80.

Gokul Hariharan

Yes. Yes.

Chia-Chang Kou

They need a low power, because our controller, Silicon Motion controller, our SM8000A, is ideal to fitting very well. This controller needs just a DDR4, or LPDDR4. So well-positioned into fitting this searching. So there is some NAND maker also want to work with us, but we have limited resources. We just have to make sure the firmware is ready. We will support two different NAND, by Q1 next year, and hopefully ramping smoothly in second half of next year.

Gokul Hariharan

As long as it is M.2 standard, you can start shipping. There is no additional qualification by individual server customer, et cetera, which takes time.

Chia-Chang Kou

We are already working with our module maker, our customer, and they will support the leading tier one server maker. They may ship earlier by Q2 next year. It is only with support controller. Our own solution, total solution, will probably be ready shipping in second half next year.

Gokul Hariharan

Understood. Second question is for Jason. Jason, first on gross margins. Historically, we have seen Ferri and any NAND solution business operate in the probably lower than 40% or 40% kind of gross margin levels. Now, obviously, you have made a meaningful step up in gross margins for solution business as well. As NAND price starts to stabilize or not really go up a lot, how should we think about the margins? Are these the steady state margins that you expect the solution business to earn, given it is more enterprise related? And secondly, could you also talk a little bit about next year looks like, based on Wallace's comment, the SSD solutions business will grow faster than 40%? What is the support that you need from an inventory perspective? Inventory already has gone up to about $800 million or so.

Could you talk a little bit about how you think about how much working capital support you would need to support that kind of additional probably 50% plus growth in the solution business?

Jason Tsai

Yes. So in terms of the gross margin question on the solutions business, as Wallace pointed out, we are negotiating 6-month pricing contracts with our customers, and we expect to turn our inventory here within that period as well. Maintaining very tight controls around working capital and price protection and margin protection for our products and our P&L. We anticipate that, certainly with automotive, these are qualified products going into projects that are multi-year, 3 years to 5 years in length. Pricing as well as supply agreements are agreed upon at the beginning of these engagements. And so we are confident that we can maintain better margins, especially as the solution complexity, the design, the support becomes more onerous. And so having the higher margin supports all of that.

In terms of the working capital, as I said, with the raise that we have done, we are confident that we can support the next couple of years' worth of business based upon what current forecasts look like. So we will continue to grow inventory, but obviously in anticipation of a 40% plus year, we have grown inventory now. We will continue to grow inventory in the next quarter, in the fourth quarter, and then we will start seeing obviously the turns of that inventory come back into cash flow as we progress.

Operator

Next question comes from the line of Daniel Yen of Morgan Stanley.

Daniel Yen

Congratulations on a great result. I want to switch gears to a very different question. Wallace, you just mentioned about some opportunities, for example, for your boot drive, for example, enterprise SSD. But I want to ask what will be your strategy right now in China. I think China is still the important region, and I think in China, we also have CSPs, we have NAND supplier, NAND makers. We also have a lot of module houses there. But at the same time, I think competition is still there. So I want to consult you on Silicon Motion's strategy right now in China both with the NAND makers, with the competitors, and also how should we address maybe the future demand from the CSP in China.

Chia-Chang Kou

I think that we really want the balance between China and also the rest of world. China has YMTC, NAND maker, they are very strong. They grow very fast. We also support and collaborate with the YMTC directly or indirectly with the module maker. I think the enterprise portion, we see they will become stronger in 2027. I think we definitely work closely for China CSP for hyperscaler. Believe us, and I think that is why we try to grow China local R&D. Eventually, we want China, our local R&D can support China customer. In Taiwan, we support the rest of world, especially for U.S. customers. And secondly, I think the China, they are growing enterprise controller company, but in the past, they also like to make enterprise C drive. In the past, as you know, NAND pricing is cyclical.

They have a downside, they have upside. So some enterprise controller company, they buy the NAND in the downside, they are selling the drive in the upside. But nowadays, there is no longer the cyclical price going up or stable, right? So very hard for enterprise controller company, you use it converting to selling the whole solution if you do not win hyperscaler or even NAND maker of the CSP customer. So I think our strategy is very simple. We winning NAND maker business and designing directly. We win hyperscaler customer in both U.S. and China. We also win CSP directly. But we do not make enterprise solution. We only sell controller and our firmware. We enable the module maker and the NAND maker. They sell and utilize our controller and the firmware, make enterprise solution, and selling to the customer. So this is where we are. We think -- we -- that is why building Silicon Motion ecosystem, working together and supporting the growing opportunity in both China and U.S.

Daniel Yen

I see. Very clear. Wallace, if I may follow up, how should we think about the overall competitive landscape on a controller in China right now? Just want to get your opinion.

Chia-Chang Kou

So for enterprise, we really do not even. There is one or two basic company move to the IPO, but we do not see they are strong enough to compete in the long term, not just one technology or business model, right? I do not really see any threat from them. Even some also want to go overseas. Today, we feel pretty comfortable for our position and with our technology, and we have a very strong developing pipeline for both our PCIe Gen 5 as well as Gen 6.

Daniel Yen

I see. Yes. The second question from me is that since right now I think you are issuing the convertible bond, I want to ask if there is any change on your dividend policy potentially in the future? Also I think just overall, how should we think about your capital structure going forward? Do you have a target, any ratio you are looking at? I think this question is actually coming from some of the investors who subscribe your convertible bond, yes.

Jason Tsai

Yes. Dividend and capital allocation certainly continues to be a priority. We increased our dividend a couple of weeks ago from $2 to $2.30 per ADS per year. We have had a history of increasing our dividend over the last 10 years or so, and our expectation is that we can continue to look at that strategically over time and continue to look at opportunities to raise that. The 3 priorities for us on capital return is dividend, share buybacks opportunistically, and then M&A as it makes sense, and that is unchanged. But certainly right now, working capital needs continue to increase as well. The CB raise puts us in a much better position to drive growth in our business with better working capital as well.

Operator

Thank you. We have now reached the end of the question and answer session. I will now turn the conference back to Mr. Wallace Kou for closing comments.

Chia-Chang Kou

Thank you everyone for joining us today and for your continuing interest in Silicon Motion. We will be attending several investor conferences over the next few months. The schedule of these events will be posted on the investor relationship section of our corporate website, and we look forward to speaking with you at these events. Goodbye for now.

Operator

Thank you for your participation in today's conference. This does conclude the program. You may now disconnect your lines.