高塔半導體 (TSEM) 2026 Q2 法說會逐字稿

Good day, and thank you for standing by. Welcome to the Tower Semiconductor second quarter 2026 earnings conference call and webcast. (Operator Instructions)

大家好，感謝您稍候。歡迎參加 Tower Semiconductor 2026 年第二季財報電話會議與網路直播。(接線員指示)

Please be advised that today's conference is being recorded. I would now like to hand the conference over to your first speaker today, Noit Levy. Please go ahead.

請注意，今天的會議將被錄音。現在我想把會議交給今天的第一位講者 Noit Levy。請開始。

Thank you. Hello, everyone, and thank you for joining us. Welcome to Tower Semiconductor's second quarter of 2026 financial results conference call. With us today are Mr. Russell Ellwanger, Chief Executive Officer; and Mr. Oren Shirazi, Chief Financial Officer.

謝謝。各位好，感謝各位加入我們。歡迎參加 Tower Semiconductor 2026 年第二季財務業績電話會議。今天與會者包括執行長 Russell Ellwanger 先生，以及財務長 Oren Shirazi 先生。

Before we begin, please note that certain statements made during today's call may be forward-looking and are subject to risks and uncertainties that could cause actual results to differ materially. These risks are detailed in our SEC filings, Form 20-F and 6-K as well as filings with the Israeli Securities Authority, all available on our website. Tower assumes no obligation to update forward-looking statements.

在開始之前，請注意今天會議中的某些陳述可能屬於前瞻性陳述，並受風險與不確定性影響，可能導致實際結果與所述內容有重大差異。相關風險已在我們向美國證券交易委員會（SEC）提交的文件（Form 20-F 與 6-K）以及向以色列證券管理局提交的文件中詳述，皆可於我們網站查閱。Tower 不承擔更新前瞻性陳述之義務。

Our second quarter 2026 results are prepared in accordance with US GAAP. Some data presented may include non-GAAP financial measures as defined under SEC Regulation G. Reconciliation to GAAP figures and full explanations are provided in today's press release and financial tables. For your reference, a supporting slide deck is available on our website and integrated into this webcast.

我們 2026 年第二季的業績係依據美國通用會計準則（US GAAP）編製。部分呈現的資料可能包含依 SEC Regulation G 定義之非 GAAP 財務衡量指標。與 GAAP 數字的調節以及完整說明已載於今日新聞稿與財務表格中。供您參考，支援用投影片簡報已上傳至我們網站，並整合於本次網路直播中。

With that, I'd like to turn the call over to our CEO, Mr. Russell Ellwanger. Russell?

接下來，我想把電話會議交給我們的執行長 Russell Ellwanger 先生。Russell？

Hello, everyone. Thank you for joining our call today. I'm truly excited to share with you the status, progress and future outlook potentials for Tower. The second quarter was quite significant, setting substantial company records across all key metrics.

各位好。感謝各位今天加入我們的電話會議。我非常興奮能與各位分享 Tower 的現況、進展以及未來展望的潛力。第二季相當重要，在所有關鍵指標上都創下公司重大紀錄。

These results continue to validate the growing value of our technology portfolio and the powerful operating leverage embedded in our business model. Our revenue is on a fervent growth trajectory with an accelerated flow through into earnings.

這些結果持續驗證我們技術組合日益提升的價值，以及我們商業模式中所蘊含的強大營運槓桿。我們的營收正處於強勁成長軌道，並以加速的速度轉化為獲利。

Second quarter revenue was $460 million, with a particularly positive profitability, 30% gross margin, 20% operating margin and 20% net margin, all being company records, excluding nonrecurring accounting items and representing, respectively, 58%, 55% and 55% quarter-over-quarter contribution from the increased revenue. These results stand as the first step of continual margin expansion we expect over the next years, driven by market-leading customer partnerships, which dictate a very rich product mix backed by strong operational execution.

第二季營收為 4.60 億美元，獲利能力表現特別亮眼：毛利率 30%、營業利益率 20%、淨利率 20%，在排除一次性會計項目後皆為公司紀錄；且相較前一季營收增加所帶來的貢獻度分別為 58%、55% 與 55%。這些成果是我們預期未來數年持續擴張利潤率的第一步，動能來自市場領先的客戶合作關係，帶來非常豐富的產品組合，並由強勁的營運執行力支撐。

Looking ahead, we guide the third quarter of 2026 midrange revenue to be $520 million, representing an annualized revenue run rate of above $2 billion. We began the year stating that Tower will have a very strong second half as the previously announced capacity investments become qualified and converted into shipments.

展望未來，我們對 2026 年第三季的營收指引中位數為 5.20 億美元，折算年化營收運行率（run rate）將超過 20 億美元。我們在年初即表示，隨著先前宣布的產能投資完成認證並轉化為出貨，Tower 下半年將非常強勁。

Beginning the second half of 2026 with a $2 billion run rate turns the page into multiple new exciting chapters for the company. The strength of our customer demand, our growing partnerships, our proven execution capabilities and the strategic investments that we have and still continue to announce provide a powerful catalyst to accelerate our short, mid- and long-term growth.

以 20 億美元的運行率開啟 2026 年下半年，意味著公司將翻開多個令人振奮的新篇章。客戶需求的強勁、合作關係的深化、我們經驗證的執行能力，以及我們已宣布且仍持續宣布的策略性投資，共同形成強大催化劑，加速我們短、中、長期的成長。

Due to direct and growing customer demand representing the scale of the market opportunity and more importantly, our ability to capture it, we have updated our 2028 model to be $3.6 billion in revenues, $1.63 billion in gross profit or 45% gross profit margin and $1.2 billion in net profit or 33% net profit margin.

由於客戶需求直接且持續成長，反映市場機會的規模，更重要的是反映我們掌握該機會的能力，我們已更新 2028 年模型：營收 36 億美元、毛利 16.3 億美元（毛利率 45%），以及淨利 12 億美元（淨利率 33%）。

Please see slides 4 and 5. Very important, the profitability gains we delivered in the second quarter are not a onetime achievement, but rather just an initial step towards profitability expansion and cash generation as represented in our updated 2028 model.

請參閱投影片第 4 與第 5 頁。非常重要的是，我們在第二季交出的獲利提升並非一次性的成就，而只是邁向獲利擴張與現金創造的起步，這也反映在我們更新後的 2028 年模型中。

Equally important, as can be seen in slides 4 and 5, this growth is accompanied by even greater efficiency. Today, we operate at about a 10 percentage point difference between gross margin and operating margin, a highly efficient structure for a company investing strongly in future growth.

同樣重要的是，如投影片第 4 與第 5 頁所示，這項成長伴隨著更高的效率。目前我們的毛利率與營業利益率之間約有 10 個百分點的差距；對一家正大力投資未來成長的公司而言，這是一個高度有效率的結構。

As revenue expands, we drive greater efficiency with operating expenses as a percentage of revenue, lowering to approximately 7%, 30% lower than current levels. This improvement is not the result of limiting investment in R&D to the exact opposite. The model includes an increase of R&D investment by over 40% against present levels. Rather, this improvement reflects the enhanced efficiency achieved through our support functions as we grow scale.

隨著營收擴大，我們將透過營業費用占營收比重下降至約 7% 來提升效率，較目前水準降低約 30%。這項改善並非來自限制研發投資，恰恰相反。該模型包含研發投資較目前水準增加逾 40%。相反地，這項改善反映的是我們在規模擴張過程中，支援性職能所達成的效率提升。

Speed and execution are one of the primary differentiators for high-tech success. We must continue to invest in R&D. This is a core capability. But as well, there must be a focus to streamline context, all of the SG&A functions.

速度與執行力是高科技成功的主要差異化因素之一。我們必須持續投資研發。這是一項核心能力。同時，也必須聚焦於精簡各項情境與所有 SG&A（銷售、一般及行政）職能。

We are doing this and likely with numbers that rival or exceed the otherwise best in the industry. Efficiency, driving the speed of execution, of course, yields strong margins. But more importantly, it enables sustained business success through the speed of execution.

我們正在這麼做，而且相關數字很可能可與業界最佳者匹敵甚至超越。效率提升能推動執行速度，當然也帶來強勁的利潤率。但更重要的是，它透過執行速度，使企業得以持續成功。

Three weeks ago, we announced a dual-track 300-millimeter capacity strategic expansion in Japan for our silicon photonics, silicon germanium and advanced optical packaging capabilities, having gained the support of the Government of Japan through the Ministry of Economics Trade and Industry, METI.

三週前，我們宣布在日本進行雙軌並行的 300 毫米產能策略性擴張，用於我們的矽光子、矽鍺以及先進光學封裝能力；並已獲得日本政府透過經濟產業省（METI）的支持。

By combining Tower's specialized technology leadership and our best of the best worldwide workforce into Japan's unparalleled manufacturing expertise and quality output, its world-class research institutions and deeply committed workforce, we are building a strategic platform that will drive innovation, economic growth and semiconductor leadership for decades to come.

透過結合 Tower 的專業技術領導力與我們全球頂尖的人才，並融入日本無與倫比的製造專業與品質產出、世界級研究機構以及高度投入的勞動力，我們正在打造一個策略平台，將在未來數十年推動創新、經濟成長與半導體領導地位。

Track 1 of this dual track add significant new 300-millimeter silicon photonics capacity with full production readiness expected during the fourth quarter of 2027. It consists of repurposing the Arai facility, formerly Fab 6, for 300-millimeter silicon photonics capacity and advanced packaging capabilities and as well as maximizing the company's Fab 7 300-millimeter output in Uozu. This Track 1 is the driver for updating the 2028 business model.

此雙軌計畫的第 1 軌將新增大量 300 毫米矽光子產能，預期於 2027 年第四季達到全面量產就緒。其內容包括將 Arai 廠（原 Fab 6）改造為 300 毫米矽光子產能與先進封裝能力，並同時最大化公司位於魚津（Uozu）的 Fab 7 300 毫米產出。第 1 軌是我們更新 2028 年業務模型的主要驅動因素。

Track 2 will commence in parallel with the first track and consists of constructing an additional 300-millimeter manufacturing facility adjacent to Fab 7. This facility is expected to provide a 4x increase in our Japanese 300-millimeter manufacturing output, focusing on silicon photonics, silicon germanium and related advanced optical packaging, positioning Tower to continue to support our accelerating customer demand for emerging AI and data center applications, driving next-generation optical connectivity requirements and is planned to provide a seamless path for Tower and our customers for continued growth post 2028.

第 2 軌將與第 1 軌同步啟動，內容是在 Fab 7 旁興建另一座 300 毫米製造設施。該設施預期可使我們在日本的 300 毫米製造產出提升 4 倍，聚焦於矽光子、矽鍺及相關先進光學封裝，使 Tower 能持續支援新興 AI 與資料中心應用所帶動的加速客戶需求、推動下一代光互連需求，並規劃為 Tower 與客戶在 2028 年之後的持續成長提供無縫銜接的路徑。

Moving to our businesses. Please refer to slide 6 as referenced for Q2 revenue breakdown. Our RF infrastructure revenues for the second quarter represented 49% of corporate revenue with approximately 43% of quarter-over-quarter growth and over 140% year-over-year growth. Silicon photonics revenue itself increased by over 60% quarter-over-quarter and over 270% year-over-year, hitting a Q2 annualized run rate of over $680 million, targeting a $1 billion run rate in Q4 '26 as the previously announced capacity expansion continues to be qualified.

接著談我們的業務。請參閱投影片第 6 頁，其中列示第二季營收拆分。第二季我們的 RF 基礎設施營收占公司總營收的 49%，約占季對季成長的 43%，且年對年成長超過 140%。矽光子營收本身季對季成長超過 60%，年對年成長超過 270%，使第二季年化運行率達到超過 6.8 億美元；隨著先前宣布的產能擴充持續完成認證，我們目標在 2026 年第四季達到 10 億美元運行率。

The full ramp of wafer starts in these investments is anticipated to occur within the fourth quarter of 2026, creating a wafer start capacity over 3 times higher than the second quarter silicon photonics revenue shipments with full financial effect anticipated to be in the second quarter of 2027. After kicking off the year with several breakthrough technology milestones, the second quarter was about further propelling Tower's leadership position in silicon photonics as an added foundation to the next phase of growth. We focus on expanding capacity, advancing our technology capabilities and deepening our engagements with strategic customers aligned to their long-term road maps.

預期這些投資所帶動的晶圓開工（wafer starts）全面爬坡將於2026年第四季內完成，形成較第二季矽光子營收出貨所對應的晶圓開工產能高出逾3倍的能力；其完整財務效益預期將於2027年第二季顯現。在年初達成數項突破性的技術里程碑後，第二季的重點在於進一步推動Tower在矽光子領域的領導地位，作為下一階段成長的額外基礎。我們聚焦於擴充產能、推進技術能力，並加深與符合其長期路線圖之策略客戶的合作。

The data center industry is undergoing a fundamental transformation as AI performance is no longer defined solely by compute. It is increasingly determined by how efficiently data moves between processors. High bandwidth, low latency, energy-efficient optical connects have become a critical enabler of AI infrastructure. While geographically distributed deployment has become equally important to hyperscalers and the communities, these trends align directly with Tower's core strengths.

資料中心產業正經歷根本性的轉型，因為AI效能已不再僅由運算能力所定義。它愈來愈取決於資料在處理器之間移動的效率。高頻寬、低延遲、節能的光學互連已成為AI基礎設施的關鍵推動者。在地理分散式部署對超大規模業者與社群同等重要之際，這些趨勢與Tower的核心優勢直接契合。

As AI clusters scale from thousands to hundreds of thousands of XPUs, electrical interconnects are rapidly approaching their practical limits. Silicon photonics has emerged as a leading platform for 800G and 1.6T pluggable optical interconnects, which have, for the most part, already replaced copper for scale-out connections outside the rack.

隨著AI叢集從數千個擴展到數十萬個XPU，電互連正迅速逼近其實務極限。矽光子已成為800G與1.6T可插拔式光學互連的領先平台，而這些互連在很大程度上已取代機架外擴展（scale-out）連線所使用的銅纜。

The next frontier is enabling optical interconnects for scale up, either within a single rack or across multiple racks and once again, silicon photonics with Tower is well positioned to lead this transition with several near package optics, NPO deployments planned over the next year and many more in design.

下一個前沿是為擴展向上（scale-up）的光學互連提供支援，可能在單一機架內或跨多個機架；而再次地，結合Tower的矽光子具備領先此一轉型的良好定位，未來一年規劃有多項近封裝光學（near package optics，NPO）部署，且仍有更多專案在設計中。

NPO delivers much greater bandwidth density and reduced energy per bit compared to pluggable optics, and yet it leverages the same established ecosystem as pluggable, overcoming the reservation from hyperscalers and data center operators about reliability, serviceability or multisource flexibility.

相較於可插拔式光學，NPO可提供更高的頻寬密度並降低每比特能耗，同時仍可沿用與可插拔式相同的既有生態系，因而化解超大規模業者與資料中心營運商對可靠性、可維修性或多來源彈性方面的疑慮。

We announced customer contracts representing approximately $1.3 billion of silicon photonics revenue for 2027 with even higher growth for 2028. With what we've stated about Track 1 investment, substantial additional 300-millimeter capacity will be added throughout 2027.

我們宣布的客戶合約代表2027年約13億美元的矽光子營收，且2028年的成長更高。依據我們對Track 1投資的說明，2027年全年將新增大量的300毫米產能。

This added 300-millimeter capacity is already requested by and committed to several lead customers, reflecting their confidence in Tower's ability not only to scale manufacturing capacity rapidly but to continue to execute on highly differentiated technology road maps required for future networking architectures.

這新增的300毫米產能已由多家主要客戶提出需求並承諾採用，反映其對Tower不僅能快速擴大量產製造產能、亦能持續執行未來網路架構所需之高度差異化技術路線圖的信心。

To support the long-term growing demand, we announced the Track 2, a most significant expansion of our manufacturing footprint in Japan, which is expected to more than quadruple its 300-millimeter capacity, positioning Tower well to support not only today's [workhorse] pluggable optics and our rapidly growing near package optics, but also the additional future market for co-packaged optics.

為支援長期成長的需求，我們宣布Track 2，這是我們在日本製造版圖最重大的擴張，預期可使其300毫米產能增加至四倍以上，使Tower具備良好定位，不僅能支援當前的（主力）可插拔式光學與快速成長的近封裝光學，也能支援未來共同封裝光學（co-packaged optics）的新增市場。

Capacity alone, however, is not enough. Our future growth will also be driven by the next generation of enabling technologies that are already moving through development pipeline. Over the next one to two years, we expect several of these technologies to transition into high-volume manufacturing, including and especially heterogeneous integration of [35] materials on silicon photonics for integrated lasers, advanced modulators and optical signal processing.

然而，僅有產能仍不足夠。我們未來的成長也將由下一代賦能技術所驅動，而這些技術已在研發管線中推進。在未來一到兩年內，我們預期其中數項技術將轉入高量產製造，包含且尤其是將[35]材料異質整合至矽光子上，用於整合式雷射、先進調變器與光訊號處理。

In support of this road map, we entered into a multiyear epitaxial wafer supply agreement with IQE securing a strategic supply of [35] epitaxial material while continuing to internalize key manufacturing steps that enhance both performance and supply chain control.

為支援此一路線圖，我們與IQE簽訂多年期外延晶圓供應協議，確保[35]外延材料的策略性供應，同時持續將可提升效能並強化供應鏈掌控的關鍵製程步驟內部化。

Our long-standing collaboration with Marvell reached an important milestone as the number of SiPho-based coherent optical modules crossed over into multimillion. This achievement demonstrates Tower's ability to manufacture some of the industry's most complex silicon photonics [ICs] at high-volume production scale.

我們與Marvell的長期合作達成重要里程碑：採用SiPho的相干光模組數量已突破數百萬。此一成就展現Tower具備以高量產規模製造業界最複雜矽光子[ICs]之一的能力。

Looking ahead, coherent photonics is expected to play an increasingly important role in scale across AI architectures where multiple data center campuses operate as a single AI factory while distributing power, cooling and infrastructure requirements across geographically diverse locations.

展望未來，相干光子預期將在AI架構的擴展中扮演愈來愈重要的角色；在此情境下，多個資料中心園區可作為單一AI工廠運作，同時將電力、散熱與基礎設施需求分散至地理位置多元的區域。

Our silicon germanium business continues to benefit from growing demand for low latency, low power efficient analog drivers and transimpedance amplifiers across traditional pluggables as well as linear pluggable optics and linear (inaudible) optics architectures. Our 100G per lane and 200G per lane products are in high-volume production across all 3, 200-millimeter fabs as we advance towards near package optics as well as 400G per lane solutions requiring tighter electrical IC, photonics IC integration and codesign. We're also seeing strong customer pull for our next-generation 300-millimeter silicon germanium platforms.

我們的矽鍺業務持續受惠於低延遲、低功耗且高效率的類比驅動器與跨阻放大器需求成長，涵蓋傳統可插拔式、線性可插拔式光學以及線性（聽不清）光學架構。我們的每通道100G與每通道200G產品已在全部3座200毫米晶圓廠進入高量產，並在推進近封裝光學以及每通道400G解決方案；後者需要更緊密的電性IC與光子IC整合與共同設計。我們也看到客戶對下一代300毫米矽鍺平台的強勁拉動需求。

Our RF mobile revenue represented 12% of our second quarter corporate revenues. As we discussed last quarter, our RFSOI business is undergoing a strategic transition from 200-millimeter to 300-millimeter manufacturing, enabling higher performance, greater integration and stronger value for our customers. In addition, we are consolidating 300-millimeter RFSOI manufacturing to Fab 7, freeing up -- I'm sorry, to Fab 10, freeing up Fab 7 capacity for a rapidly growing SiPho and silicon germanium business. These factors resulted in a 14% decrease in our 300-millimeter year-over-year RFSOI revenues.

我們第二季的RF行動通訊營收占公司總營收的12%。如同我們上季所討論，我們的RFSOI業務正進行從200毫米轉向300毫米製造的策略性轉型，以提供更高效能、更高整合度並為客戶創造更高價值。此外，我們正將300毫米RFSOI製造整併至Fab 7，釋放出——抱歉，是整併至Fab 10，釋放Fab 7產能以支援快速成長的SiPho與矽鍺業務。上述因素導致我們300毫米RFSOI營收年對年下降14%。

Looking forward, we have realized a very strong design win momentum for our 300-millimeter platform, particularly for premium smartphones. A road map repeat with best-in-industry figures of merit has gained market excitement and engagement, driving an expected 3 times RFSOI increase and 300-millimeter wafer starts by mid-2027 against the Q2 '26 shipments.

展望未來，我們在300毫米平台上已實現非常強勁的設計導入（design win）動能，特別是在高階智慧型手機領域。具備業界最佳性能指標（figures of merit）的路線圖重複採用（repeat）已引發市場高度關注與參與，推動預期到2027年年中，RFSOI與300毫米晶圓開工將較2026年第二季出貨增加3倍。

Power management revenue for the second quarter represented 14% of corporate revenues with year-over-year revenue growth and strong demand for both our 200-millimeter and 300-millimeter BCD offerings. Our technology focus on power delivery for high-performance computing, gives us a leadership position in load gate charge and low Rdson, LDMOS devices. Our latest generation power technology enables our customers to develop high frequency, high efficiency, DC to DC converters for a variety of growth segments.

第二季電源管理營收占公司總營收的14%，營收年對年成長，且我們200毫米與300毫米BCD產品皆需求強勁。我們聚焦於高效能運算的供電技術，使我們在負載閘極電荷（load gate charge）與低Rdson的LDMOS元件方面具備領導地位。我們最新一代電源技術使客戶能為多個成長領域開發高頻、高效率的DC-DC轉換器。

During this quarter, we experienced increased demand from existing customers and also saw very strong new customer acquisitions across our power portfolio. This momentum is primarily driven by growth sectors where power density and thermal efficiency are mission-critical. Through close collaboration with our lead customers, we continue to advance our next-generation power management road map.

本季我們來自既有客戶的需求增加，同時也在電源產品組合上看到非常強勁的新客戶取得。此一動能主要由電源密度與熱效率至關重要的成長領域所驅動。透過與主要客戶的緊密合作，我們持續推進下一代電源管理路線圖。

Sensor display for the second quarter represented 12% of our corporate revenue. In our image sensor business year-over-year revenue is predominantly flat. However, we're seeing a sharp surge in demand, particularly in the machine vision market for high-end, high-resolution sensors used in semiconductor inspection, driven by the accelerated build-out of DDR and HBM memory assembly lines. And as well in the automotive industry, especially for EV battery inspection.

第二季感測器顯示業務占公司總營收的12%。我們的影像感測器業務營收年對年大致持平。然而，我們看到需求明顯激增，特別是在機器視覺市場，對用於半導體檢測的高階高解析度感測器需求強勁，這主要由DDR與HBM記憶體組裝產線加速擴建所帶動。汽車產業亦然，尤其是電動車電池檢測。

This strong demand is expected to continue to grow over the next two years. We are well positioned to support it with a range of products our lead customers have developed on our state-of-the-art global (inaudible) 300-millimeter platform.

預期這股強勁需求在未來兩年將持續成長。我們具備良好定位可提供支援，憑藉主要客戶在我們最先進的全球（聽不清）300毫米平台上所開發的一系列產品。

Looking at utilization. During a period of high capacity ramp, Fab 2, Fab 3 and Fab 9, 200-millimeter Fabs operated utilization rates between 80% and 85%. Fab 5 in Japan was at 75% utilization. Fab 7 continues to be fully utilized, well above our 85% utilization model.

就稼動率而言。在產能快速爬坡期間，Fab 2、Fab 3與Fab 9等200毫米晶圓廠的稼動率介於80%至85%。日本的Fab 5稼動率為75%。Fab 7仍持續滿載運轉，明顯高於我們85%稼動率的模型假設。

Now I'd like to turn the call to our CFO, Mr. Oren Shirazi. Please, Oren. Thank you.

現在我想把電話交給我們的財務長 Oren Shirazi 先生。請，Oren。謝謝。

Thank you. Hello, everyone. Earlier today, we released our financial results for the second quarter of 2026. I will now review the highlights of these results as well as the balance sheet CapEx investments and our updated business model.

謝謝。各位好。今天稍早，我們已發布 2026 年第二季的財務結果。接下來我將回顧這些結果的重點，以及資產負債表、資本支出（CapEx）投資與我們更新後的商業模式。

First, looking into the P&L. Revenue for the second quarter of 2026 was $460 million, a record in the company's history, representing 11% quarter-over-quarter growth compared to $414 million in the first quarter of 2026 and 24% year-over-year growth compared to $372 million in the second quarter of 2025. Gross profit for the second quarter of 2026 was a record $138 million, reflecting a 30% gross margin and an increase of 72% compared to $80 million in the second quarter of 2025.

首先看損益表。2026 年第二季營收為 4.60 億美元，創公司歷史新高，較 2026 年第一季的 4.14 億美元季增 11%，較 2025 年第二季的 3.72 億美元年增 24%。2026 年第二季毛利為 1.38 億美元，同樣創新高，毛利率 30%，較 2025 年第二季的 8,000 萬美元增加 72%。

Operating profit for the second quarter of 2026 was $90 million, 2.26 times the operating profit in the second quarter of 2025. Net profit for the second quarter of 2026 was $91 million, reflecting a 20% net margin, and an increase of 95% or $44 million compared to a net profit of $47 million in the second quarter of 2025. Earnings per share for the second quarter of 2026 were $0.80 per share basic and $0.79 diluted, almost double the $0.49 basic and $0.41 diluted in the second quarter of 2025.

2026 年第二季營業利益為 9,000 萬美元，為 2025 年第二季營業利益的 2.26 倍。2026 年第二季淨利為 9,100 萬美元，淨利率 20%，較 2025 年第二季 4,700 萬美元的淨利增加 95%，或增加 4,400 萬美元。2026 年第二季每股盈餘為基本每股 0.80 美元、稀釋每股 0.79 美元，幾乎是 2025 年第二季基本每股 0.49 美元、稀釋每股 0.41 美元的兩倍。

Moving to our balance sheet. As we previously announced, we received $290 million in prepayments from SiPho customers in the first quarter of 2026, mostly towards 2027 capacity reservation. These customer prepayments are included in the balance sheet as of the end of June 2026 as liabilities under short- and long-term customer advances and are included in the cash flow report for Q1 '26 and for H1 '26 as cash from operating activities.

接著看資產負債表。如我們先前宣布，2026 年第一季我們收到 SiPho 客戶 2.90 億美元的預付款，主要用於 2027 年產能預訂。截至 2026 年 6 月底，這些客戶預付款在資產負債表中列為負債，歸類於短期與長期的客戶預收款；並在 2026 年第一季及 2026 年上半年（H1 '26）的現金流量表中列為營業活動之現金流入。

Our balance sheet continues to be strong, evidenced by the following indicators and financial ratio. As of the end of June 2026, our assets totaled $3.8 billion, primarily comprised of $1.3 billion in net fixed assets, predominantly fair machinery and $2 billion of current assets. Our current ratio remains very strong at about 4.9 times, while shareholders' equity reached a record of $3.1 billion at the end of June 2026.

我們的資產負債表持續強勁，以下指標與財務比率可為佐證。截至 2026 年 6 月底，我們資產總額為 38 億美元，主要由 13 億美元的固定資產淨額（以廠務設備與機台為主）以及 20 億美元的流動資產構成。我們的流動比率維持在約 4.9 倍的強勁水準，而股東權益於 2026 年 6 月底達到創紀錄的 31 億美元。

Hedging. I would like now to describe our currency hedging activities. Regarding the Japanese yen, since the majority of TPSCo's revenue is denominated in yen, and the vast majority of TPSCo's cost are also in yen, we have a natural hedge over most of our Japanese business and operations. To mitigate part of the remaining yen exposure, we execute zero cost cylinder transactions to hedge currency fluctuations. And while the yen exchange rate against the dollar may fluctuate, there is limited impact on our margin.

避險。我現在想說明我們的匯率避險活動。就日圓而言，由於 TPSCo 的大部分營收以日圓計價，且其絕大多數成本也以日圓計價，因此我們在日本業務與營運的大部分具有自然避險效果。為降低剩餘的部分日圓曝險，我們會執行零成本區間（zero cost cylinder）交易，以對沖匯率波動。因此，即使日圓兌美元匯率可能波動，對我們毛利率的影響也有限。

Regarding the Israel shekel, while we have no revenue in this currency, a portion of our cost in Israel is denominated in shekel. So we also hedge a large portion of that currency risk by entering into zero cost cylinder transactions to mitigate this exposure. And while the shekel exchange rate against the US dollar may fluctuate, the impact on our margin is limited as seen over the past few quarters, during which the shekel appreciated strongly against the US dollar.

就以色列謝克爾而言，雖然我們沒有以此貨幣計價的營收，但我們在以色列的一部分成本以謝克爾計價。因此，我們也透過簽訂零成本區間（zero cost cylinder）交易，對沖其中很大一部分的匯率風險，以降低曝險。而即使謝克爾兌美元匯率可能波動，對我們毛利率的影響仍然有限；從過去幾季可見，期間謝克爾兌美元大幅升值。

Now moving to our CapEx investment plan. As previously announced, in order to support the increasing SiPho and SiGe demand, we are executing a $920 million investment plan to expand capacity and capability of SiGe and SiPho equipment across our eight-inch plants in Israel, Newport Beach and Texas as well as our 12-inch Uozu fab in Japan.

接下來談我們的資本支出（CapEx）投資計畫。如先前宣布，為支援持續增加的 SiPho 與 SiGe 需求，我們正在執行一項 9.20 億美元的投資計畫，以擴充我們位於以色列、Newport Beach 與德州的 8 吋廠，以及日本魚津（Uozu）12 吋晶圓廠的 SiGe 與 SiPho 設備產能與能力。

This investment is on track in terms of purchase orders issued, technology and process qualification, equipment arrivals and ramp plans. Approximately 50% of this $20 million CapEx investment has been paid to date, and this is included in our cash flow for investing activities for the reporting period through the second quarter of 2026, while the remaining 50% is expected to be paid during H2 '26 and full year 2027.

就採購單發出、技術與製程認證、設備到廠以及量產爬坡計畫而言，該投資進度皆符合預期。截至目前，這項 2,000 萬美元的資本支出投資約有 50% 已支付，並已反映在截至 2026 年第二季的本報導期間投資活動現金流中；其餘 50% 預計將於 2026 年下半年（H2 '26）及 2027 全年支付。

Business model. As recently announced and presented earlier today by Russell, the company updated its business model, which now includes the planned investment in the Arai facility, formerly Fab 6, to repurpose it for 12-inch SiPho wafer manufacturing, thereby maximizing the company's 12-inch wafer output in Japan supported by METI, Japan's Ministry of Economic Trade and Industry.

商業模式。如近期宣布，並由 Russell 於今天稍早的簡報中所呈現，公司已更新其商業模式；其中納入對 Arai 廠區（前 Fab 6）的規劃投資，將其改造為 12 吋 SiPho 晶圓製造，以在日本最大化公司 12 吋晶圓產出，並獲得日本經濟產業省（METI）的支持。

The additive model is based on forward-looking operational business and financial assumptions, including the assumption that all fabs will operate at 85% utilization post the full installation and qualification of the $920 million in SiPho and SiPho CapEx investment. We continue to execute to ramp up SiPho and SiGe capacity and capability in our Fab 2, 3, 5 and 7, and post the planned repurposing of the Arai facilities and qualification of the new equipment tools to be installed there.

此加成模型（additive model）係基於前瞻性的營運、業務與財務假設，包括假設在 9.20 億美元 SiPho 與 SiPho 資本支出投資完成全面安裝與認證後，所有晶圓廠將以 85% 的稼動率運作。我們持續推進在 Fab 2、3、5 與 7 的 SiPho 與 SiGe 產能與能力爬坡，以及在 Arai 廠區完成規劃改造後，並對將於該處安裝的新設備機台進行認證。

Another assumption is the assumption regarding forward-looking wafer selling prices for existing and future products and flow. And an assumption, forward-looking assumptions regarding cost and completion date for the successful qualification of tools to be installed, process technologies to be qualified and customer products to be qualified and ordered from the company.

另一項假設是關於現有與未來產品及產線的前瞻性晶圓售價假設。以及關於將要安裝之設備機台成功認證、需完成認證的製程技術、以及需完成認證並向公司下單之客戶產品，其成本與完成日期等前瞻性假設。

Under the updated model, we target $3.6 billion in annual revenue, which is $760 million higher than the February model target, reflecting 27% revenue increase. $1.63 billion annual gross profit, resulting in a 45% gross margin as compared to 39% in the prior model. This annual gross profit is $510 million higher than the prior February 2026 model, representing 67% incremental gross profit derived from the incremental $760 million revenue, reflecting our enhanced product mix. $1.38 billion of annual operating profit, resulting in a 38% operating margin as compared to 32% in the prior model.

在更新後的模型下，我們目標為年營收 36 億美元，較 2 月模型目標高出 7.60 億美元，反映營收增加 27%。年毛利 16.3 億美元，毛利率 45%，相較於先前模型的 39%。此年毛利較先前 2026 年 2 月模型高出 5.10 億美元，代表由新增 7.60 億美元營收所帶來的增量毛利為 67%，反映我們更優化的產品組合。年營業利益 13.8 億美元，營業利益率 38%，相較於先前模型的 32%。

This annual operating profit is $480 million higher than the prior February 2026 model, representing 63% incremental operating profit derived from the incremental $760 million revenue, again, reflecting our enhanced product mix. And lastly, $1.2 billion in annual net profit, resulting in a 33% net margin as compared to 26% in the prior model and as compared to 20% in the second quarter of 2026. The final net profit is $450 million higher than the prior February 2026 model, representing 59% incremental net profit derived from the incremental $760 million of additional revenue, reflecting our enhanced product mix.

此年營業利益較先前 2026 年 2 月模型高出 4.80 億美元，代表由新增 7.60 億美元營收所帶來的增量營業利益為 63%，同樣反映我們更優化的產品組合。最後，年淨利 12 億美元，淨利率 33%，相較於先前模型的 26%，也高於 2026 年第二季的 20%。最終淨利較先前 2026 年 2 月模型高出 4.50 億美元，代表由新增 7.60 億美元營收所帶來的增量淨利為 59%，反映我們更優化的產品組合。

That concludes my prepared remarks. Now I'd like to turn the call back to the operator so we can take your questions.

以上是我準備的發言。現在我想把電話交回給接線員，以便開始問答。

(Operator Instructions)

(接線員指示)

Cody Acree, Benchmark/StoneX.

Cody Acree，Benchmark/StoneX。

Congrats on another great quarter and just great execution. Congrats, everybody. Just a point of clarification and then a couple of quick questions. Did you update the SiPho bookings number for '27, the $1.3 billion? Would you give a new version of that?

恭喜再創一個非常出色的季度，也展現了很強的執行力。恭喜各位。我先釐清一點，接著問幾個簡短問題。你們是否更新了 2027 年 SiPho 訂單（bookings）數字，也就是 13 億美元那個數字？能否提供一個更新版本？

No. No, it did not. I stated in the script that the Q4 start rate, which would be fully realized in Q2 '27 revenue was 3 times higher than the Q2 shipment. So that number is--

不。不，沒有。我在講稿中提到，第四季的起始速率（在 2027 年第二季的營收中會完全反映）是 2027 年第二季出貨量的 3 倍。所以那個數字是——

Yes, this can be [680 times 3,] but this is capacity. While the $1.3 billion is on customer committed prepayments or on customer committed contracts.

是的，這可以是 [680 乘以 3]，但這是產能。而 13 億美元是客戶承諾的預付款或客戶承諾的合約。

As stated additionally, the capacity growth is spoken for. Is it all booked? No, but it's spoken for.

另外如同所述，產能成長已經有人要了。都已經全數下單了嗎？沒有，但已經被預訂了。

Is it fair to -- I mean, a touch up?

這樣說公平嗎——我是說，稍微修正一下？

No, I was just going to state that I think there was no update given on the contracts and the bookings, but there was an update given pretty much on what is started or will be started for shipments.

不，我只是想說，我認為合約和訂單（bookings）沒有提供更新，但幾乎有更新的是已開始或將開始用於出貨的部分。

Okay. Excellent. And then just with that level of visibility, you've been able to put up some record sequential and annual growth rates both on a percentage and on a dollar basis, it's been pretty impressive. With this level of visibility, can you maybe just handicap the likelihood of being able to sustain this kind of growth rates or even accelerate from here?

好的。非常好。然後在這樣的可見度下，你們無論以百分比或以金額計，都做出了創紀錄的連續季與年度成長率，確實相當令人印象深刻。在這樣的可見度下，你能否大概評估一下，維持這種成長率、甚至從這裡再加速的可能性？

Well, we gave the 2028 financial model and stated that it is our expectation to reach those numbers at a minimum by run rate nominally in the full year in 2028. So I think you could estimate what the growth rate is off of a $3.6 billion 2028 revenue level.

嗯，我們給出了 2028 年的財務模型，並表示我們的預期是最晚在 2028 年完整年度中，以名目上的 run rate 至少達到那些數字。所以我想你可以用 2028 年 36 億美元的營收水準去推算成長率。

Okay. And then lastly, Russell. Maybe, if you can just help me to get a scale on the Japanese projects, the Track 1 and Track 2. I'm trying to understand the wafer volumes or the revenue support that ultimately will be available out of Japan, not so much Track 1.

好的。最後，Russell。也許你可以幫我掌握一下日本專案的規模，Track 1 和 Track 2。我想了解最終日本能提供的晶圓量或營收支撐，不是那麼著重在 Track 1。

I think that's clear in your fiscal '28 model that, that is assuming full utilization of that Track 1 build-out. Is that the right way to think about it?

我想在你們 2028 財年的模型裡已經很清楚，那是假設 Track 1 擴建能滿載利用。這樣理解對嗎？

Yes, at 85% utilization, correct.

是的，以 85% 的稼動率，沒錯。

Okay. And so then how do we think about the scale of Track 2?

好的。那我們該如何看待 Track 2 的規模？

We stated that the Track 2 would quadruple the 300-millimeter capacity and that is predominantly for SiPho and SiGe. We didn't give specific numbers. Part of the reason for not giving numbers is that we're in final negotiations, strong negotiations, not on the pricing part of it, just on the timing part to complete the facility and complete the facilitization of the facility and then the negotiations with the suppliers on the equipment itself.

我們說過 Track 2 會讓 300 毫米產能增加到 4 倍，而那主要用於 SiPho 和 SiGe。我們沒有給出具體數字。部分原因是我們正處於最後談判階段、而且是強力談判——不是在價格上，而是在時程上：完成廠房、完成廠房的設施化（facilitization），以及與供應商就設備本身進行談判。

So budgets, everything is approved, everything is there. We're focused on covering everything with internal cash creation. So it sits nicely in our hands. We're not looking at anything being gated or leveraged by a fundraising event or anything of that sort. There's no dilution within the plans.

所以預算等一切都已核准，一切都到位。我們專注於用內部現金創造來涵蓋所有支出。因此這完全掌握在我們手上。我們不會讓任何事情取決於募資事件或類似事項，也不會被其所限制或槓桿化。計畫中沒有稀釋。

But we don't yet, at this point, have the final schedule of tool installations and tool qualification. Our target is that everything is installed and functioning by Q4 2028. And as stated in the script, that it then provides a seamless growth trajectory into 2029, having reached the full potential of the Track 1 development in 2028.

但目前我們還沒有設備安裝與設備認證（qualification）的最終時程。我們的目標是在 2028 年第四季前讓一切安裝完成並可運作。而且如同講稿所述，在 2028 年達到 Track 1 開發的全部潛力後，這將在 2029 年提供無縫的成長軌跡。

So our target would be to have that grow from there. The exact numbers of how much we would get from that, that we haven't said, and it's not necessarily our desire to give quarterly updates and financial models.

所以我們的目標是讓它從那裡繼續成長。至於我們能從中得到多少的確切數字，我們沒有說，而且我們也不一定希望每季提供更新與財務模型。

But you could expect that within the first quarters as we get into 2027 that we'll update our long-term financial model, depending upon how it actually turns out on the final build-out of the factory and the amount of tools that we can fit into the built-out factory. So some of that, even some of the building plan is not yet finalized.

但你可以預期，當我們進入 2027 年的前幾個季度時，我們會視工廠最終擴建的實際結果，以及我們能在擴建後的工廠中放入多少設備，來更新我們的長期財務模型。所以其中一些內容，甚至部分建築規劃，目前都尚未定案。

Okay. Excellent. And then just to be clear, the 4 times quadrupling of the capacity of Track 2, that is just the Uozu facility? Or is that quadrupling accounting Fab 6?

好的。非常好。然後再確認一下，Track 2 產能增加到 4 倍，是只指魚津（Uozu）廠嗎？還是這個 4 倍也包含 Fab 6？

We're including Fab 6 in that 4 times number. So whatever we would be doing incrementally in Fab 6, which predominantly we'll be using as an epicenter and also for specialty packaging tools. But yes, the Fab 6 incremental that you see in Track 1 is included into the 4 times.

我們把 Fab 6 也包含在這個 4 倍的數字裡。所以我們在 Fab 6 會做的任何增量，主要會把它用作核心據點（epicenter），也用於特殊封裝設備。但沒錯，你在 Track 1 看到的 Fab 6 增量也包含在這個 4 倍之中。

Mehdi Hosseini, SIG.

Mehdi Hosseini，SIG。

Yes. I do have a couple. Russell, I just wanted to better understand the evolving end market demand. As we look into next year, and NPO becomes material, does that give you ability to increase your content per given transceiver?

是的。我有幾個問題。Russell，我想更了解終端市場需求的演變。展望明年，當 NPO 變得重要時，是否會讓你們有能力提高每個收發器的內容價值（content）？

To the extent I understand your question, I would say no. It basically would be -- we would be selling an NPO in those cases rather than selling pluggable, and in many instances, most likely be selling both. But the content itself, no, at least to the extent that I understand your question.

就我對你問題的理解，我會說不會。基本上會是——在那些情況下我們會賣 NPO，而不是賣可插拔（pluggable），而且在很多情況下，很可能兩者都會賣。但就內容本身而言，不會，至少就我對你問題的理解是如此。

Sure. I'm trying to better understand if there is synergy here, especially as you -- I'm under assumption that at some point, the [PIC] itself will require a stacking of SiGe and SiPho. And that's where the question is originating from, if we have more of a heterogeneous structure, a PIC-based structure, would that -- I imagine that would be more positive for you, and I just want to see if I'm in the right frame of mind.

了解。我想更了解這裡是否有協同效應，尤其是——我假設在某個時間點，[PIC] 本身會需要 SiGe 與 SiPho 的堆疊。所以我的問題源自於此：如果我們有更多異質整合的結構、以 PIC 為基礎的結構，那是否——我想像那對你們會更有利，我只是想確認我的思路是否正確。

I think you're in the correct frame of mind, and I think the answer is yes. But I believe that, that focus is more not on NPL, but on next generation. So when we talked about the advanced packaging, a good amount of the advanced packaging that we're putting in place is die to wafer and wafer to wafer bonding.

我認為你的思路是正確的，而且我認為答案是肯定的。但我相信那個重點更多不在 NPL，而是在下一代。所以當我們談到先進封裝時，我們正在導入的先進封裝有相當一部分是晶粒對晶圓（die to wafer）與晶圓對晶圓（wafer to wafer）鍵合。

Okay. And I imagine that would also require you to increase SiGe capacity and that's already embedded in the Phase 1 of the capacity expansion, right?

好的。我想像那也會需要你們增加 SiGe 產能，而這已經包含在第一階段（Phase 1）的產能擴張中，對嗎？

To some extent, it's mainly embedded into the Phase 2. I'm sorry, Phase 1 is focused very, very strong directly on SiPho expansion where we, as stated, all of that growth that we're doing in Phase 1 is already spoken for.

在某種程度上，但主要是包含在第二階段（Phase 2）。抱歉，第一階段非常、非常強烈地聚焦在 SiPho 擴產；如同所述，我們在第一階段所做的所有成長都已經有人要了。

Okay. Great. And then second question, regarding your manufacturing footprint, given your emphasis in Japan and how you have this Phase I and Phase 2. Should I assume that majority of your U.S.-based customers would be supported through facilities in Japan or U.S. would -- or is there another area where you eventually have to increase capacity?

好的。很好。第二個問題，關於你們的製造版圖，鑑於你們強調日本，並且有 Phase I 和 Phase 2。我是否應該假設，你們多數美國客戶會由日本的設施來支援？或者美國——或者是否還有其他地區，最終你們必須再增加產能？

We have and we continue to increase capacity in both Newport Beach and in San Antonio, and we're still planning to increase further capacity in both sites. But one of the reasons for having chosen Japan, sorry, is its geopolitical neutrality.

我們已經、而且也持續在 Newport Beach 與 San Antonio 兩地擴充產能，我們仍計畫在兩個據點進一步擴產。但選擇日本的原因之一——抱歉——是它在地緣政治上的中立性。

There's no issues for somebody to be supplied out of Japan. So it's a very good place to grow. I mean there's multiple reasons to one that we chose Japan. We have truly a remarkable workforce there. Most of the R&D that we have in Japan was a derivative of Panasonic hires back when Panasonic was the creme de la creme. I mean a top three Japanese company and hiring their choice of people, predominantly from Tokyo University. So we still have an extremely, extremely capable core workforce as far as R&D.

對於從日本供應產品給任何人而言都沒有問題。所以那是一個非常適合成長的地方。我的意思是，我們選擇日本有很多原因。我們在那裡確實擁有一支非常卓越的勞動力。我們在日本的大部分研發人員，都是當年從松下延攬而來的衍生團隊，當時松下可說是業界的翹楚。我的意思是，它是日本前三大的公司之一，能招募到他們最優秀的人才，主要來自東京大學。因此，就研發而言，我們至今仍擁有一支極其、極其有能力的核心團隊。

Additionally, anybody in the world that's ever been around Japanese manufacturing, the quality of it is just outstanding. It's part of a Japanese mentality. I mean, ancillary here, but we lived in Japan for a number of years when our youngest daughter was going through kindergarten.

此外，世界上任何接觸過日本製造業的人都知道，它的品質就是非常出色。這是日本人心態的一部分。順帶一提，我們在日本住了好幾年，當時我們最小的女兒正在上幼兒園。

And it's a very, very interesting thing. Already at 3.5, 4 years of age, when a kindergarten student goes to school, part of their uniform is a plastic container with a wash cloth in it called an oshibori, and the school day starts by taking out that washout and wiping down their desk. And the school day ends by taking out that wash cloth and wiping down their desk.

而這是一件非常、非常有趣的事。早在3.5歲、4歲時，幼兒園孩子上學時，制服的一部分是一個裝著擦手巾的塑膠盒，叫做 oshibori（濕毛巾），而一天的課程從拿出那條毛巾擦拭自己的桌子開始。而一天的課程結束時，也會拿出那條毛巾再把桌子擦一遍。

The honoring of the workplace is a very, very big, big thing in Japan. And that culture when you started 3.5, 4 years of age, learning to honor the workplace, that stays with you forever. I mean I did a film once, so it's kind of funny, just about the procedure of returning your tray and dishes after lunch. And the activities in Japan, it's more or less of an assembly line. There's absolutely no degradation of a quality mentality versus when they're in the regular factory. So when it deals with very, very high-quality manufacturing, I think Japan is really the best of the best.

在日本，尊重工作場所是一件非常、非常重要的事。而當你從3.5歲、4歲就開始學習尊重工作場所，這種文化會伴隨你一輩子。我的意思是，我曾經拍過一段影片，所以有點好笑，內容只是關於午餐後歸還餐盤和餐具的流程。而在日本，這些活動幾乎就像一條組裝線。他們在這方面的品質心態，和在正式工廠裡時完全沒有任何落差。所以談到非常、非常高品質的製造，我認為日本真的是最頂尖中的最頂尖。

So those are multiple reasons for going there. It was the taxonomy that we have of workforce is fantastic. Quality culture in Japan is fantastic. Education in Japan is amazingly good. And it's geopolitically neutral.

所以去那裡有很多原因。我們的人才結構非常出色。日本的品質文化非常出色。日本的教育好得驚人。而且在地緣政治上相對中立。

Got it. And just a very quick follow-up. When should RF mobile revenue stabilize? Your prepared remarks suggest that you're continuing to consolidate move manufacturing to 300-millimeter. Would those changes in your manufacturing footprint stabilize into the second half? Or is this something that is going to carry on into '27?

了解。再非常快速追問一下。RF 行動（mobile）營收何時會趨於穩定？您在事先準備的發言中提到，您仍在持續整併並把製造移轉到 300 毫米。這些製造版圖的變動會在下半年趨於穩定嗎？還是這會一路延續到 2027 年？

I wouldn't use the term stabilize, stabilize would mean going into an area where you're not having incremental growth. But I believe I had stated that from the tape-outs activity, the design activity that we've already won that we would see Q2, Q3 of next year, having the largest 300-millimeter manufacturing that we've ever done for RFSOI.

我不會用「穩定」這個詞，穩定意味著進入一個不再有增量成長的區域。但我相信我曾提到，從我們已經贏得的 tape-out 活動、設計活動來看，明年第二季、第三季，我們將看到我們在 RFSOI 上有史以來最大的 300 毫米製造量。

But I wouldn't say stabilizing, it's just getting to a point of continued growth and then additional plans, how do we grow it beyond that. But yes, I would think that for the Fab 10 factory where we're manufacturing, it will be very, very full come second, third quarter of next year.

但我不會說是穩定，而是到達一個持續成長的階段，然後再規劃如何在此之上進一步成長。不過是的，我認為就我們生產的 Fab 10 工廠而言，到了明年第二、第三季會非常、非常滿載。

Richard Shannon, Craig-Hallum Capital Group LLC.

Richard Shannon，Craig-Hallum Capital Group LLC。

Excellent. Congratulations on wonderful results. Keep up the good work here. My first question is going to be a follow-on from a couple of Cody's questions here and trying to think about the ultimate silicon photonics and SiGe 300-millimeter capacity after you finish Track 2 and you use the statements of 4 times increase in capacity, I think, at least 4 times capacity.

太好了。恭喜你們取得非常出色的成果。請繼續保持。我的第一個問題是延續 Cody 的幾個問題，想請教在你們完成 Track 2 之後，最終的矽光子（silicon photonics）與 SiGe 300 毫米產能會是什麼樣子；你們提到產能將提升 4 倍，我想至少是 4 倍的產能。

So it seems like we need to understand and characterize how much of the capacity to that point is 200 millimeter versus 300 millimeter. So I was wondering if you could answer that quantitatively or at least qualitatively to help us think about that more specifically.

所以看起來我們需要理解並界定，到那個時間點為止的產能中，有多少是 200 毫米、多少是 300 毫米。因此我想請問你是否能用量化，或至少用質化的方式回答，幫助我們更具體地思考這件事。

Honestly, I'm not sure I understand the question. Could you just restate it, please?

老實說，我不確定我理解這個問題。你可以請再重述一次嗎？

Sure. You've said that after Track 2, you're going to -- it's going to increase your 300-millimeter silicon photonics and SiGe 5G capacity by -- or quadruple. So in order for us to quantify this, I think we need to understand what that position -- what that mix of capacity between 200 millimeter and 300 millimeter would be at that point before starting that capacity increase. So trying to figure out -- I wonder if you could answer what that split of capacity is at that point, so we can try to calculate that.

當然。你們說在 Track 2 之後，你們的 300 毫米矽光子與 SiGe 5G 產能將會——或是將會增加到四倍。所以為了讓我們能量化，我想我們需要了解在開始那次產能提升之前，那個時間點的產能結構——也就是 200 毫米與 300 毫米的產能組合。所以想弄清楚——我想請問你能否回答在那個時間點的產能拆分比例，讓我們可以試著計算。

Yes. I just have to (inaudible) real quick, sorry. I'll be right back. Sorry, one second, sorry. Sorry, Richard. I would say that, to begin with, after I would say, '27 mid '28, all of the growth that we have in SiGe and SiPho, and definitely in SiPho, will be in 300 millimeter.

可以。我得先（聽不清）很快一下，抱歉。我馬上回來。抱歉，等一下，抱歉。抱歉，Richard。我會說，首先，大約在 2027 年中到 2028 年中之後，我們在 SiGe 和 SiPho 的所有成長——尤其是 SiPho——都會在 300 毫米上。

The ForEx increase in capacity in Japan, the agreement with METI is really focused on optics and it's on the silicon photonics and silicon germanium. So as a minimum, we would intend to be adding 20,000, 25,000 wafer per month silicon photonics capacity, that can go much, much higher than that. And that's the desire of the company -- I mean, of the country. So if you multiply that by 2.25, you're dealing with very, very big amount of wafer capacity versus 200-millimeter.

日本 ForEx 的產能擴增，以及與 METI 的協議，重點其實是光學，聚焦在矽光子與矽鍺（silicon germanium）。所以至少，我們打算新增每月 20,000 到 25,000 片晶圓的矽光子產能，而且還可以再高很多、很多。而那也是公司——我是說，國家——所希望的方向。所以如果你把那個數字乘以 2.25，相較於 200 毫米，你談的是非常、非常龐大的晶圓產能。

So although I'm not sure why it's so important for you to know that ratio to model something. The growth in silicon photonics is quite big and the predominant portion coming out of Track 1, almost 100% of it will be silicon photonics. You have the delta numbers in revenue from the previous model to the present model.

所以雖然我不太確定為什麼你需要知道那個比例來建模。矽光子的成長相當大，而從 Track 1 出來的主要部分，幾乎 100% 都會是矽光子。你們也看得到從先前模型到目前模型在營收上的增量數字。

And then when we go into the 300-millimeter facility, the actual split, we've not yet announced, but the predominant portion of everything of that 4 times growth will be in silicon photonics. And on the next biggest portion of it will be in silicon germanium.

然後當我們進入 300 毫米廠時，實際的拆分我們尚未公布，但那 4 倍成長中的絕大部分都會是矽光子。其次最大的部分會是矽鍺。

Okay. I'll just state also, I think everyone is trying to figure out what the next update on the business model looks like. I think you said you'll update us kind of earlier or some point next year here. So we're all trying to do that work ahead of time. So we'll look forward to more detail when you're able to give it.

好的。我也想補充一下，我想大家都在試著推估下一次商業模型更新會長什麼樣子。我想你說過會在明年較早的時間或某個時間點更新給我們。所以我們都在提前做這些工作。因此也期待你們在方便時提供更多細節。

My follow-on -- one of my follow-on questions here is on NPL. And I'd love to get a sense of the degree to which this will be a meaningful contributor in SiPho revenues in 2027. Is this going to be -- how would you characterize the size either quantitatively or qualitatively, please?

我的追問——其中一個追問是關於 NPL。我很想了解，這在 2027 年對 SiPho 營收會在多大程度上成為有意義的貢獻者。這會是——你會如何用量化或質化來描述它的規模，謝謝？

I think it will be significant. The exact percent, I don't yet know. But I would think it will be in the -- not in the single digits, but in the tens of the percentage of what we'll be shipping, especially in the second half of the year.

我認為這將會很顯著。確切的百分比我目前還不知道。但我認為會是——不是個位數，而是我們出貨量的百分之十幾到數十幾，尤其是在下半年。

Okay. Great. That's very helpful. And then my last question is on advanced packaging here. I think there's a little bit of contribution in the Track 1 investments in Japan, but I think a bigger part of Track 2. So I'm going to ask a little bit longer term question here, Russell, which is I think in the fulsome of time here, after everything is built and how you think the market is going to evolve, how much of your silicon photonics revenues are going to be packaging related?

好的。很好。這非常有幫助。接著我最後一個問題是關於先進封裝。我認為在日本的 Track 1 投資中有一些貢獻，但我認為 Track 2 的占比更大。所以我想問一個稍微偏長期的問題，Russell：我想在時間充分推進、等一切都建好之後，並且看你如何看待市場演進，你們的矽光子營收中有多少會與封裝相關？

That's a very good question. I'm not sure how much of it necessarily is packaging related versus being packaging enabled. So even right now for the integrated laser, for an indium phosphide laser, we would see next year several tens of millions of revenue from integrated laser. The integrated laser is a chip to wafer bonding.

這是個非常好的問題。我不確定其中有多少必然是「封裝相關」，還是說是「由封裝所促成」。所以即使是現在，對於整合式雷射、以磷化銦雷射而言，我們明年會看到整合式雷射帶來數千萬美元的營收。整合式雷射是晶片到晶圓的鍵合。

That's packaging. It's right now outsourced, but we'll be bringing that in-house. So it's not that we're getting paid per se for packaging, and we're not trying to compete with packaging houses. We're bringing more capability in-house that we have control over the end result and certainly much more control over the start to ship time of wafers rather than to depend on the supplier no matter how good they are.

那就是封裝。目前是外包的，但我們會把它導入內部。所以並不是說我們會因封裝本身而收費，我們也不打算與封裝廠競爭。我們把更多能力內製，是為了能掌控最終結果，當然也能更好地掌控從開始到晶圓出貨的時間，而不是不論供應商多優秀都仍需依賴對方。

But we are not looking at per se competing in packaging. We're looking at packaging being an enabler to grow our core business at the highest quality, best speed to production of any supplier that there would be. Hopefully, that answers your question, Richard. We're not going to separate a packaging revenue. We're not focused on it as a packaging revenue rather than as an enabler for our silicon photonics platform.

但我們並不是要去競爭封裝本身。我們把封裝視為推動核心業務成長的使能因素，讓我們能以最高品質、最快速度量產，優於任何供應商。希望這回答了你的問題，Richard。我們不會把封裝營收單獨拆出來。我們的重點不是把它當作封裝營收，而是把它當作我們矽光子平台的使能因素。

And now we'll go and take our next question, and the question comes from of Cody Acree from Benchmark/StoneX.

接下來我們進入下一個問題，提問者是來自 Benchmark/StoneX 的 Cody Acree。

Quick follow-up, guys. Russell, with all the capacity additions that have been happening around the industry from some of your peers at the GLOBALFOUNDRIES.

快速追問一下，各位。Russell，隨著產業內一些同業（包括 GLOBALFOUNDRIES）近期都在擴增產能……

Can you pick up on the first part? I apologize. Could you start again?

你能重複一下前半段嗎？不好意思。可以再說一次嗎？

It's fine. With all the capacity additions around the industry from some of your peers, GLOBALFOUNDRIES, STMicro, Samsung have all made announcements about planned capacity additions. If you can look out 12, 18, 24 months, can you maybe just frame your opinion of the supply/demand health of the industry with all of these different tranches coming in line?

沒問題。隨著產業內一些同業的產能擴增，GLOBALFOUNDRIES、STMicro、Samsung 都已宣布規劃中的產能擴充。如果把時間拉到 12、18、24 個月後，你能否大致說明一下，在這些不同批次的產能陸續到位之際，你對產業供需健康狀況的看法？

Supply is certainly increasing. We have a definite benefits of anyone right now, and that we're qualified at the lead customers worldwide with very strong contracts lasting through '28. And at this point, given additional capacity coming online, strong interactions and discussions to maintain contracts well beyond that, I don't necessarily have too good of a feeling for how much added capacity will be coming into the market. I've honestly not followed up on that so strongly. Our focus has really just been how we maintain full share or majority share of our lead customers, and we're in very good shape there.

供給確實在增加。我們目前有一個明確的優勢：我們已在全球的領先客戶端完成認證，且有非常強的合約一路延續到 2028 年。而在此時點，考量到新增產能上線，以及為了把合約維持到更久之後所進行的密切互動與討論，我不一定對會有多少新增產能進入市場有很好的掌握。老實說我沒有那麼密切地追蹤。我們的重點其實一直是：如何在領先客戶那裡維持全部或多數的市占，而我們在這方面狀況非常好。

The most important thing is something I talked about within the script, and that is speed. So as long as -- we have programs going on for not just next generation, but generation plus two in many cases, generation plus 3. Those programs always enable you to come to the market faster, stronger than anyone else.

最重要的是我在講稿裡提到的一點：速度。所以只要——我們不僅有下一代的計畫，在很多情況下還有下一代加二，甚至下一代加三的計畫。這些計畫總是能讓你比任何人更快、更強勢地把產品推向市場。

Tied to that, what's very, very real from a business standpoint, if we're working with a very large customer, and we're putting substantial resources into next-generation or multiple generation of future developments, there's typically exclusivity agreements on both sides. And we would enable a lead customer to have a head start, especially for any module that's joint developed and we would request 100% market share.

與此相關，從商業角度來看有一件非常、非常真實的事：如果我們與非常大型的客戶合作，並投入大量資源在下一代或多代的未來開發上，通常雙方都會有排他性協議。我們會讓領先客戶取得先發優勢，特別是對於共同開發的任何模組，而我們會要求 100% 的市占。

So I don't -- again, for part of your question, I can't -- I don't really have a good feel for how much capacity is additionally coming into the market. What I think we have a much better feel on is the market share that we'll be maintaining with our lead customers and we believe that will stay extremely high.

所以我不——再說一次，關於你問題的一部分，我無法——我確實不太掌握還會有多少額外產能進入市場。但我認為我們更有把握的是：我們將在領先客戶那裡維持的市占，而我們相信那會維持在極高水準。

And Russell, is there -- just for my own edification, a figure of merit stratification, I guess, that you can point us to that would be a good reference point so we can just keep an eye on your continued leadership in the industry?

Russell，那是否——只是為了我自己更了解，有沒有一個你能指引我們的「績效指標」分層（figure of merit stratification）之類的參考點，讓我們可以持續關注你們在產業中的領先地位？

The figure of merit that you're referring to is what? I mean, typically, a figure of merit is a technical achievement. What are you looking for? You're looking for something to measure our leadership?

你所指的 figure of merit 是什麼？我的意思是，通常 figure of merit 是技術成就的指標。你在找什麼？你是在找一個用來衡量我們領先地位的東西嗎？

No, I guess I'm just trying to gauge some of the industry rhetoric about differing capabilities from your different competitors around the industry. And I'm just trying to see if there's something you can help us with to help us better understand where it is that Tower really differentiates.

不是，我想我是在試著判斷產業裡一些說法，談到你們不同競爭對手之間能力差異的那些論述。我只是想看看你能否提供一些協助，讓我們更好理解 Tower 到底在哪些地方真正有差異化。

Sorry, I don't want this to sound facetious, but we differentiate in figure of merit. Insertion loss is probably one of the biggest things that one could look for. And if you have at this point a best-in-breed insertion loss, you really help the integrator, number one, by not needing to buy more expensive CW lasers because of greater output, but also the ability to reduce the amount of lasers that's in the package.

抱歉，我不希望這聽起來像是在開玩笑，但我們的差異化就在 figure of merit 上。插入損耗（insertion loss）可能是最重要、最值得看的指標之一。如果你在此時點擁有業界最佳的插入損耗，你就能在兩方面幫助系統整合商：第一，不需要因為要更高輸出而去買更昂貴的連續波（CW）雷射；第二，也能減少封裝內所需的雷射數量。

We have press released before. It was a press release with InnoLight about our insertion loss being an enabler to have the amount of [CW] lasers that they need in the package. So those are -- I believe was having it. I have to go back to the PRs a couple of years ago. But that becomes really the biggest differentiator that we can have is to lead the industry a figure of merit.

我們之前發過新聞稿。那是一則與 InnoLight 的新聞稿，談到我們的插入損耗如何成為使能因素，讓他們能在封裝中配置所需數量的 [CW] 雷射。所以那些——我相信是有的。我得回去翻一下兩年前的新聞稿。但這確實成為我們能擁有的最大差異化：在 figure of merit 上領先業界。

If you're looking at next-generation modulators at the OFC conference, we did a joint PR with Coherent about being able to have done a 400G modulator in silicon. And there is no secret that we and maybe some others are working with thin film lithium niobate. We also are working with indium phosphide for a modulator. So the best way that one could be looking at maintaining market share is, how close are you in working with the next generation and or two generations out? And how is your performance in those areas. So does that answer your question, Cody? Hopefully, it does.

如果你看 OFC 會議上的下一代調變器，我們與 Coherent 發了一則聯合新聞稿，談到能在矽上做出 400G 調變器。而且也不是秘密：我們以及可能還有一些其他公司正在研究薄膜鈮酸鋰（thin film lithium niobate）。我們也在用磷化銦來做調變器。所以，觀察能否維持市占的最佳方式是：你在下一代、以及往後一到兩代的研發合作上有多接近？以及你在那些領域的效能表現如何。所以這有回答你的問題嗎，Cody？希望有。

It does. That was very helpful. And then lastly, you did mention InnoLight. I don't know if you've had a chance, I know you're busy this morning, but to see some of the press reports that the administration is maybe looking to limit exports of Chinese technology into the U.S. from an optical standpoint. Just wondering if you have any thoughts there?

有的。這非常有幫助。最後，你提到了 InnoLight。我不知道你今天早上是否有機會看到一些媒體報導——我知道你很忙——報導說政府可能正考慮從光學角度限制中國技術出口到美國。想請問你對此有沒有什麼看法？

Many thoughts, but nothing that I would want to say publicly.

有很多想法，但沒有任何我想公開說的。

And we're going to take our final question for today, and the question comes line of Lisa Thompson from Zacks Investment Research.

我們將進行今天最後一個提問，問題來自 Zacks Investment Research 的 Lisa Thompson。

I just have a couple of -- two questions. First off, are you experiencing any shortages or supply chain issues for your own production? I know you were worked on the indium phosphate issue.

我只有幾個——兩個問題。首先，你們自有生產是否正面臨任何短缺或供應鏈問題？我知道你們一直在處理磷化銦的問題。

Big picture, no. We're in very good position on across the board with starting materials and with what we would call variable materials that are needed to manufacture. We were under somewhat of a crunch for indium phosphide starting material.

從大方向來看，沒有。在起始材料以及我們所稱製造所需的可變材料方面，我們整體狀況非常好。我們曾在磷化銦起始材料方面有些吃緊。

And we believe that we have addressed that very nicely with the contract with IQE. But other than indium phosphide, which we had several ways that we went after to increase the amount of substrates during a difficult period. And I think that, that we've gotten resolved as well. No, we're in very good shape on supply.

我們相信，透過與 IQE 的合約，我們已非常妥善地解決了這個問題。但除了磷化銦之外，在一段困難期間，我們也透過好幾種方式來增加基板供應量。我認為那也已經解決了。沒有，我們的供應狀況非常良好。

Okay. And then my last question is if we're going to do, say, an error analysis on your business model, where do you think the most variability of outcomes is? Is it going to be expenses or timing or the prices you forecast you'll get for your products? Where is the risk?

好的。那我最後一個問題是：如果要對你們的商業模式做一個誤差分析，你認為結果變動性最大的地方在哪裡？會是費用、時程，還是你們預估產品可取得的價格？風險在哪裡？

Yes. I think I addressed it in my prepared remarks that we are based on a few assumptions which are important. One of them is the selling price per wafer. The second is the cost assumptions, the time of installation and qualification. And the third one is that we will utilize 85% of the Fab. And this is the basic assumptions for the model, I can say.

是的。我想我在事先準備的發言中已提到，我們是基於幾個重要假設。其中之一是每片晶圓的售價。第二是成本假設、安裝與認證（qualification）的時間。第三是我們將使用晶圓廠（Fab）85% 的產能。我可以說，這些就是模型的基本假設。

Which has the biggest range of outcome, which has the biggest range of outcome that could be the most or to look at?

哪一項的結果區間最大、最值得關注？

Per wafer because if you have more or less quantity, so you have more or less variable costs associated with that. But if the price goes up, it goes all the way to the bottom line and vice versa. So maybe usually, but it's a general statement. So usually, the selling price is just 100% reflection over the margin.

每片晶圓的售價，因為如果你的數量多或少，就會有相應多或少的變動成本。但如果價格上升，就會直接反映到最終盈餘，反之亦然。所以通常可能是這個，但這只是一般性的說法。因此通常，售價對毛利的反映是 100%。

Dear speakers, there are no further questions for today. I would now like to hand the conference over to Russell Ellwanger for any closing remarks.

各位講者，今天沒有其他問題了。現在我想把會議交給 Russell Ellwanger 做結語。

Firstly, as I started the call, I'll end it with the same statement. I am extremely excited with where we're at, what we're doing our future prospects, enjoyed being able to share them with you.

首先，正如我在電話會議開始時所說的，我也用同樣的話作結。我對我們目前所處的位置、我們正在做的事以及未來前景感到非常振奮，也很高興能與各位分享。

One of the most exciting things about being involved in high-tech business and maybe in particular, being in management or the CEO is the fact that in high-tech, normally, your interfacing, interacting with groupings of people that are from any statistical standpoint in the upper end of intelligence. And that's a wonderful thing, a wonderful place to be dealing with. When you couple that with people a very high character, it becomes fantastic, and it's wonderful, wonderful interaction.

參與高科技產業最令人興奮的事情之一——或許特別是身處管理層或擔任 CEO——在於在高科技領域，你通常會與一群從任何統計角度來看都屬於高智商上端的人互動、合作。這是一件很美好的事，也是很棒的環境。當再加上這些人品格非常高尚時，就變得非常精彩，互動也非常、非常美好。

I can really state that if I look at across the board in business, but right now, specifically on silicon photonics, every one of our customers that we have there with really out exception are just outstanding people that we deal with. And every interaction, it's more than joyful, it's a partnership to where you have open communication, you work with each other.

我可以很坦白地說，放眼整個商業界，但就目前而言，特別是在矽光子方面，我們在那裡的每一位客戶幾乎無一例外都是我們合作起來非常出色的人。每一次互動都不只是愉快，更像是一種夥伴關係：你們開放溝通、彼此合作。

And off of that, you go into the next generation. We've had a very good call the other day with a big customer with regard to having now announced -- well, not announced, but having a very strong additional 300-millimeter capability. And the partnership feeling the way that we're growing together, it's really an amazing place to be.

而在此基礎上，你就會進入下一代。前幾天我們與一位大客戶進行了一次非常好的通話，內容是關於我們現在已經宣布——嗯，並非正式宣布，但已具備非常強的額外 300 毫米產能。那種夥伴關係、以及我們一起成長的感覺，真的是一個令人驚嘆的處境。

And one of my really greatest joys about leading Tower is the ability that I have daily inside the company to be dealing with very smart people that are of high character and being able to interact with customers that are very smart, of high character and seeing the integration of both together towards making new things.

而我領導 Tower 最大的喜悅之一，是我每天在公司內能與非常聰明且品格高尚的人共事，也能與同樣非常聰明、品格高尚的客戶互動，並看到雙方整合在一起去創造新事物。

We've talked multiple times and it was in the script as well, not just of Tower's growth, but of Tower and our customers' growth. and looking at things in that regard in that respect, it's a fantastic way to be. So I truly -- if we look at the financial model, if we look at achieving a 33% net profit, just very, very thrilled about where we're at, where we're going. And all of these opportunities in front of us. The really an extremely wonderful adventure that we're entering into Japan to build out hub of optical excellence for the SiPho, for the silicon germanium.

我們多次談到——腳本中也有提到——不只是 Tower 的成長，也包括 Tower 與我們客戶的成長。從這個角度來看，這是一種非常棒的狀態。所以我真的——如果我們看財務模型、看達成 33% 淨利率，我對我們目前的位置與前進方向感到非常、非常振奮。以及擺在我們面前的所有機會。我們即將在日本展開的旅程，去打造矽光子（SiPho）與矽鍺（silicon germanium）的光學卓越樞紐，真的是一段極其精彩的冒險。

We're really at an amazing place that everyone, not that we haven't in the past, but people come to work now even more excited than they did before. There's so much happening. And that's an amazing place to be. So in stating that, my invitation to everyone that's an investor, reach out, we'd love to talk to you. We'd love to have as much interaction as we can, help you better understand what we're doing, where we're at, and be able to share with you the strong results that we're having in the company.

我們真的處在一個令人驚嘆的階段；並不是說過去沒有，但現在大家來上班比以前更興奮。有太多事情正在發生。這是一個很棒的狀態。因此，我想邀請所有投資人主動聯繫我們，我們很樂意與各位交流。我們希望能有盡可能多的互動，幫助各位更了解我們在做什麼、我們目前在哪裡，並與各位分享公司正在取得的強勁成果。

That being said, we look forward to engaging with the investment community at the upcoming Jefferies conference. August 25, 26 in Chicago, and at the Benchmark StoneX Annual Conference in New York on September 10. The invitation for interactions with investors, it's real. And the better you understand our story, the better you understand where we're at, where we're going, we think the best for everybody.

話說回來，我們也期待在即將到來的 Jefferies 會議上與投資界互動。8 月 25、26 日在芝加哥，以及 9 月 10 日在紐約舉行的 Benchmark StoneX 年度會議。與投資人互動的邀請是真誠的。你越了解我們的故事、越了解我們目前的位置與未來方向，我們認為對大家都越好。

So with that, I'd like to close and just thank you for your interest and for your support. Thank you. Bye-bye.

那麼我就此結束，並感謝各位的關注與支持。謝謝。再見。

This concludes today's conference call. Thank you for participating. You may now all disconnect. Have a nice day.

今天的電話會議到此結束。感謝各位參與。各位現在可以掛線。祝各位有美好的一天。