Cadence（CDNS）— 2026-10-26 3Q26 法說會逐字稿（英文原文＋中譯）

日期：2026-10-26（美國時間盤後） 逐字稿為 Cadence Design Systems earnings call 音檔 AI 轉寫，英文原文與中譯並列；專有名詞與數字以公司 press release／10-Q 為準。

校正紀錄

本檔為 AI 語音轉寫，落地前已校正下列轉寫誤植；未列出者為原文。Memo 區塊與 Transcript 區塊同步修正。 - 「genetic workflow／基因工作流程」→ agentic workflow（代理式工作流程） - 「Atop Telco」→ a top telco（一家頂級電信商，未具名）；前後文為「與全球知名系統客戶」「與頂級電信商」兩項並列合約，Memo 區塊同步 - 中譯 hyperscalers 誤作「超級計算機」「超級擴展者」→ 超大規模雲端業者；semi-customers／「半定製客戶」→ 半導體客戶 - Memo 區塊「標記效率」→ token 效率（token efficiency） - 「Linux expansion／Linux 擴展」→ logo expansion（新客戶擴展） - 「I agree with that, Andrew」→ Anirudh（講者稱呼誤植）；「Siddhi」→ Siti（Siti Panigrahi，Mizuho） - Joe Vruwink 問答中 CEO 稱「Renesas reported a 34X productivity improvement with ViraStack」→ InnoStack（準備稿明載 Renesas MCU 團隊部署 InnoStack 達 34 倍，與同段 Samsung Foundry、Socionext、Rapidus 皆屬 InnoStack 一致） - Gary Mobley 提問稱「bookings in the first quarter ... around 26% year over year」：期間無從判斷（疑為第三季或年初至今），保留原文並標 [期間存疑]，Memo 區塊同步；CFO 回答未引用該數字 - 「CAITs」（CFO 答 Lee Simpson）全名無從判斷，保留原文並標 [待確認] - 「Hexagon D&E ... initially expected to contribute about $160 million」：口述未指明為營收或營業利益，標 [待確認] - 財務四項交叉驗算：①GAAP 營業利益率 26.2%＜非 GAAP 44.1% ✓；2026 展望：營收中點 64.02 億 ÷ (1+21%) ＝ 2025 約 52.9 億，營收增加約 11 億 ✓（與 CFO「約 11 億」一致）；非 GAAP 營益率中點 44.7%×64.02 億＝約 28.6 億，較 2025 增加約 5 億 ✓；②全年非 GAAP EPS 中點 8.40，4Q 中點 2.22、3Q 2.11，隱含 1H26 合計約 4.07 美元†，合理；GAAP EPS 4.84–4.90 對非 GAAP 8.37–8.43 差距合理 ✓；③各產品成長（EDA +20%／IP +24%／SDNA +30%）加權≈整體 +22% ✓；④Rule of 40：21%＋44.7%＝65.7%，與「north of 65%」✓；4Q 營收指引中點 17.10 億，3Q 16.33 億，QoQ 約 +4.7%† - 「TSMC 預測五年內晶片尺寸成長 48 倍」為 CEO 轉述，單位口徑（面積或電晶體數）未明，保留原話 - 「SDNA」為 CEO 口述之產品群縮寫（疑 SD&A，System Design & Analysis），保留原文

警語：此 Memo 和逐字稿為 AI 生成，可能因為音檔品質導致轉錄逐字稿有誤，或是 AI 曲解原意／幻覺導致 Memo 內容有誤，請以原始音檔為準。

Memo 亮點 • Cadence 交出另一個優異的季度，所有關鍵指標均遠超預期。訂單積壓達到創紀錄的 82 億美元，公司將全年營收展望上調至 21% 的成長（根據語音資料推斷）。每股盈餘展望提高 30 美分，達到 8.40 美元。 [CEO - 00:02:16] • 第三季營收年增 22%，總營收為 16.33 億美元。GAAP 營業利潤率為 26.2%，非 GAAP 營業利潤率為 44.1%。GAAP 每股盈餘為 1.05 美元，非 GAAP 每股盈餘為 2.11 美元。 [CFO - 00:18:11] • 經常性營收年增 18%。核心 EDA 營收成長 20%，IP 成長 24%，SDNA 成長 30%。 [CEO - 00:10:05] • Agentic AI 與 Maestro 領域專用 AI 模型展現強勁的早期客戶吸引力，已有多項部署進行中。初步結果顯示，品質提升高達 30%，且相較於通用基礎模型，token 效率提升至 5 倍。 [CEO - 00:04:22] • 硬體（Palladium 與 Protium 仿真平台）交出另一個創紀錄季度，主要受 AI 與高效能運算客戶前所未有的需求推動。 [CEO - 00:12:49] • 董事會已授權額外 25 億美元用於股票回購計畫。第三季已使用 4 億美元回購股票。 [CFO - 00:19:14] 地緣政治 • 更新後的展望假設目前實施的出口管制規定在今年剩餘時間內將保持大致不變。 [CFO - 00:19:36] 產業循環週期 • 類比與混合訊號領域動能強勁，受物理 AI 及整體市場復甦推動。記憶體相較一兩年前展現顯著改善。 [CEO - 00:43:00] 產業供需 • 設計活動加速持續推動核心 EDA、SDA 及 IP 產品的強勁需求。Agentic AI 產品組合亦促進基礎產品需求增加。 [CEO - 00:10:19] 產業狀況 • Cadence 在其 Agentic AI 產品組合上看到顯著的客戶動能，推動基礎產品需求增長。隨著 AI 與高效能運算設計規模與複雜度提升，硬體輔助驗證需求強勁。 [CEO - 00:10:29] 獲利 • 第三季 GAAP 營業利潤率為 26.2%，非 GAAP 營業利潤率為 44.1%。GAAP 每股盈餘為 1.05 美元，非 GAAP 每股盈餘為 2.11 美元。 [CFO - 00:18:38] • 根據更新的 2026 年展望中點，Cadence 預期營收成長 21%，營業利潤率 44.7%，每股盈餘 8.40 美元（根據語音資料推斷）。 [CFO - 00:21:42] • 40 法則指標：根據指引，Cadence 預期約 21% 的營收成長及略低於 45% 的利潤率，2026 年合計將超過 65%（根據語音資料推斷）。 [CEO - 00:30:35] 公司展望、未來成長動能 • Cadence 將全年營收展望上調至 21% 成長，並將每股盈餘展望提高 30 美分至 8.40 美元（根據語音資料推斷）。公司具備良好條件，能把握 Agentic AI 的轉型機會。 [CEO - 00:02:37] • Cadence 預期 2026 年經常性營收將持續表現優異，主要受 Agentic AI 解決方案及基礎工具使用增加推動。 [CEO - 00:28:21] • 環境對 Cadence 持續非常有利，AI 各方面設計活動充沛，幾乎所有業務部門均取得強勁市佔率提升。 [CEO - 00:31:22] • Cadence 正重新定義 Agentic AI 在各產業的應用，推出自有領域專用模型，同時保持與所有前沿模型合作的彈性。 [CEO - 00:31:56] 市場競爭策略 • Cadence 在幾乎所有業務領域均取得市佔率提升，包括 IP、硬體、EDA 及 Agentic AI。產品組合為歷來最強。 [CEO - 00:31:41] • Maestro 不限於 Cadence 自有模型，保持最大彈性以使用最適合各任務的模型，包括前沿模型。Cadence 目標成為客戶主要介面，確保資料機密性並提供雲端或本地部署彈性。 [CEO - 00:40:32] 公司對市場觀點的看法 • 《華爾街日報》在 2026 年最佳董事會名單中，將 Cadence 董事會評為標普 500 公司中排名第二。 [CEO - 01:04:41] 營收 • 第三季營收為 16.33 億美元，年增 22%。2026 年營收預計介於 63.87 億至 64.17 億美元之間。第四季預期營收為 16.95 億至 17.25 億美元。 [CFO - 00:18:32] • 年營收成長率：核心 EDA 20%，IP 24%，SDNA 30%。 [CEO - 00:10:05] 訂單 • 訂單積壓達到創紀錄的 82 億美元，反映出因客戶需求增長而帶動的強勁訂單。 [CEO - 00:02:27] • 第一季[期間存疑]訂單約為中間值 20% 區間，年增約 26%。儘管 2026 年為低續約期，但強勁的訂單積壓與訂單量由健康的設計活動及擴大客戶合作推動。 [Analyst - 00:57:14] 產品組合 • 所有產品群組均交出優異成績。核心 EDA、IP 及 SDNA 均呈現顯著年增。Agentic AI 產品組合獲得強勁吸引力。 [CEO - 00:09:56] 營收地區 • 本季經常性營收成長受中國營收變動影響，與去年第三季相比，今年未出現類似變動。 [CEO - 00:27:50] 下游客戶 • Cadence 與頂尖客戶簽署多項重要且廣泛的協議，包括與全球知名系統客戶的廣泛擴展及一家頂級電信商（a top telco），確立 Cadence 為其主要設計夥伴。 [CEO - 00:10:55] • 超過 100 家活躍客戶參與 Cadence 的 AI 超級代理。Maestro 及其他 Agentic AI 解決方案的早期部署已在多家客戶中展開。 [CEO - 00:05:48] 同業 • 競爭對手宣佈 GPT-Synopsys；Cadence 強調 Maestro 的彈性，能同時使用自有及前沿模型，並具備資料機密性保障及部署彈性。 [CEO - 00:39:18] 存貨庫存 • 本季庫存再次增加，以支持客戶訂單，因 Cadence 正處理龐大訂單積壓。處理此積壓無產能或供應鏈限制。 [CFO - 00:34:19] 出貨週期與交期 • 硬體業績季度間可能波動，視產品交付及客戶安裝時程而定。基礎需求環境仍強勁且健康。 [CFO - 00:34:04] 新產品／新技術 • Cadence 推出 Maestro，一套針對晶片與系統設計的領域專用 AI 模型組合。初步結果顯示品質提升高達 30%，token 效率較通用基礎模型提升 5 倍。 [CEO - 00:04:22] • AI 超級代理（ChipStack、ViraStack、InnoStack、AuraStack）獲得強勁吸引力，客戶回報生產力提升：Renesas 34 倍、Socionext 5 倍、Rapidus 2 倍、Nokia 5 倍。 [CEO - 00:50:52] 法規影響 • 更新後的展望假設目前實施的出口管制規定在今年剩餘時間內將保持大致不變。 [CFO - 00:19:36] 成本毛利 • 2026 年非 GAAP 營業利潤率預計約為 44.7%，略高於去年。今年營收較去年增加約 11 億美元，非 GAAP 營業利潤增加約 5 億美元。 [CFO - 00:37:42] 銷售／管理／研發費用 • 第三季管理及一般費用激增，因年度慈善捐款時點不同，今年於第三季發生，去年則在第四季。 [CFO - 00:34:40] 營業利益 • 第三季 GAAP 營業利潤率為 26.2%，非 GAAP 營業利潤率為 44.1%。 [CFO - 00:18:38] 股本、增資、可轉換債券、庫藏股 • 董事會已授權額外 25 億美元用於股票回購計畫。第三季已使用 4 億美元回購股票。 [CFO - 00:19:14] 現金、現金流 • 現金餘額為 14.52 億美元，未償還債務本金為 26 億美元。第三季營運現金流為 4.24 億美元。2026 年營運現金流預計約 20 億美元，超過 50% 的自由現金流將用於股票回購。 [CFO - 00:18:52] 併購 • 儘管整合 Hexagon 的 D&E 業務並在策略成長領域進行目標投資，Cadence 預期 2026 年非 GAAP 營業利潤率約為 44.7%。Hexagon D&E 表現略優於預期，隨整合進展預計明年將為 Cadence 帶來益處。 [CFO - 00:37:42] 長短期借款 • 第三季末未償還債務本金為 26 億美元。 [CFO - 00:18:52] 應收帳款 • 第三季應收帳款天數為 66 天。 [CFO - 00:19:14] Transcript Presentation Jade - Conference Operator

(00:00:00)Good afternoon. My name is Jade, and I will be your conference operator today. At this time, I would like to welcome everyone to the Cadence Third Quarter 2026 Earnings Conference Call. All lines have been placed on mute to prevent any background noise. After the speakers' remarks, there will be a question-and-answer session. If you would like to ask a question during this time, simply press star, then the number one on your telephone keypad. Thank you. I will now turn the call over to Richard Gu, Vice President of Investor Relations for Cadence.

(00:00:00)下午好。我叫 Jade，今天將擔任您的會議操作員。此時，我想歡迎大家參加 Cadence 2026 年第三季度財報電話會議。所有電話線已被靜音，以防止任何背景噪音。在發言者發言後，將進行問答環節。如果您想在此期間提問，只需按下星號，然後按電話鍵盤上的數字一。謝謝。現在我將電話轉交給 Cadence 投資者關係副總裁 Richard Gu。

Jade - Conference Operator

(00:00:31)Please go ahead.

(00:00:31)請開始。

Richard Gu - Cadence, Investor Relation

(00:00:35)Thank you, operator. I'd like to welcome everyone to our third quarter 2026 earnings conference call. I'm joined today by Anirudh Devgan, President and Chief Executive Officer, and John Wall, Senior Vice President and Chief Financial Officer. The webcast of this call and a copy of today's prepared remarks will be available on our website, cadence.com. Today's discussion will contain forward-looking statements, including our outlook on future business and operating results. Due to risks and uncertainties, actual results may differ materially from those projected or implied in today's discussion.

(00:00:35)謝謝你，操作員。我想歡迎大家參加我們的 2026 年第三季度財報電話會議。今天我與 Anirudh Devgan，總裁兼首席執行官，以及 John Wall，高級副總裁兼首席財務官一同參加。本次通話的網絡直播和今天的準備發言稿副本將在我們的網站 cadence.com 上提供。今天的討論將包含前瞻性聲明，包括我們對未來業務和經營結果的展望。由於風險和不確定性，實際結果可能與今天討論中預測或暗示的結果有重大差異。

Richard Gu - Cadence, Investor Relation

(00:01:13)For information on factors that could cause actual results to differ, please refer to our SEC filings, including our most recent Forms 10-K and 10-Q. CFO commentary. and today's earnings release. All forward-looking statements during this call are based on estimates and information available to us as of today, and we disclaim any obligation to update them. In addition, all financial measures discussed on this call are non-GAAP unless otherwise specified. The non-GAAP measures should not be considered in isolation from or as a substitute for GAAP results.

(00:01:13)有關可能導致實際結果不同的因素，請參閱我們的 SEC 申報，包括我們最近的 10-K 和 10-Q 表格。首席財務官的評論。以及今天的財報發布。本次通話中的所有前瞻性聲明均基於截至今天可用的估算和信息，我們不承擔更新的任何義務。此外，除非另有說明，否則本次通話中討論的所有財務指標均為非 GAAP 指標。非 GAAP 指標不應單獨考慮或作為 GAAP 結果的替代。

Richard Gu - Cadence, Investor Relation

(00:01:50)Reconciliations of GAAP to non-GAAP measures are included in today's earnings release. For the Q&A session today, I would ask that you observe a limit of one question only. If time permits, you may re-queue with additional questions. Now I'll turn the call over to Anirudh.

(00:01:50)GAAP 與非 GAAP 指標的調整包含在今天的財報發布中。在今天的問答環節中，我要求您僅限於一個問題。如果時間允許，您可以重新排隊提出其他問題。現在我將電話轉交給 Anirudh。

Anirudh Devgan - Cadence, CEO

(00:02:06)Thank you, Richard. Good afternoon, everyone. And thank you for joining us today. I'm pleased to report that Cadence delivered another outstanding quarter, with all key metrics coming in well above our guidance. Backlog reached a record $8.2 billion, reflecting strong bookings driven by growing customer demand. Given our strong business momentum, expanding technology leadership, and compelling AI growth opportunities, we are raising our revenue outlook for the year to 21% growth. We are also increasing our EPS outlook by 30 cents to $8.40.

(00:02:06)謝謝你，Richard。大家下午好。感謝您今天的參加。我很高興地報告，Cadence 又交出了一個出色的季度，所有關鍵指標均超出我們的指導範圍。訂單積壓達到創紀錄的 82 億美元，反映出強勁的訂單量，這是由於客戶需求的增長。鑒於我們強勁的業務動能、擴大的技術領導地位和引人注目的 AI 增長機會，我們將今年的收入展望上調至 21% 的增長。我們還將每股收益展望上調 30 美分至 8.40 美元。

Anirudh Devgan - Cadence, CEO

(00:02:58)John will provide more details on both our Q3 results and the updated financial outlook. Agentic AI represents a transformational market opportunity, and Cadence is exceptionally well positioned to capitalize on it. Our differentiated three-layer framework is anchored by our physically accurate simulation and optimization engines in the critical middle layer, tightly integrated with AI agents and orchestration at the top, and accelerated compute and proprietary data at the base. Agentic AI continues to evolve at a remarkable pace.

(00:02:58)John 將提供有關我們第三季度業績和更新的財務展望的更多細節。Agentic AI 代表著一個變革性的市場機會，而 Cadence 在這方面的定位非常優越。我們的差異化三層框架以我們在關鍵中層的物理準確模擬和優化引擎為基礎，與頂層的 AI 代理和編排緊密集成，並在底層加速計算和專有數據。Agentic AI 仍在以驚人的速度發展。

Anirudh Devgan - Cadence, CEO

(00:03:48)With our AI super agents gaining strong traction, we are now addressing the long-standing customer demand for domain-specialized models. We believe the future of agentic AI-based design lies in increasingly domain-specialized models combined with intelligent model routing for performance and reinforcement learning loops with underlying tools for recursive self-improvement. Today, I'm excited to introduce Maestro, a portfolio of domain-specialized AI models purpose-built for chip and system design. By controlling all three layers—our AI superagents, Maestro models, and proven EDA and SDA tools—we are uniquely positioned

(00:03:48)隨著我們的 AI 超級代理獲得強勁的市場反響，我們現在正在滿足客戶對領域專用模型的長期需求。我們相信，基於 agentic AI 的設計未來將在越來越多的領域專用模型與智能模型路由相結合，以提高性能和強化學習循環，並使用基礎工具進行遞歸自我改進。今天，我很高興介紹 Maestro，一組專為晶片和系統設計而構建的領域專用 AI 模型。通過控制所有三層——我們的 AI 超級代理、Maestro 模型和經過驗證的 EDA 和 SDA 工具——我們獨特地定位

Anirudh Devgan - Cadence, CEO

(00:04:46)to deliver optimal performance, efficiency, and throughput results for our customers. Initial results are very compelling, demonstrating up to 30% higher quality and up to 5x greater token efficiency versus general-purpose foundation models. Moreover, the Maestro models can be further post-trained on customers' proprietary data to deliver even better outcomes. Our model router chooses among our models and other leading frontier and open-weight models as needed for optimal results. Customers can opt to deploy Maestro models in the cloud or in their on-premises environment.

(00:04:46)以為我們的客戶提供最佳的性能、效率和吞吐量結果。初步結果非常引人注目，顯示出比通用基礎模型高出多達 30% 的質量和高達 5 倍的令牌效率。此外，Maestro 模型可以在客戶的專有數據上進一步進行後訓練以提供更好的結果。我們的模型路由器根據需要在我們的模型和其他領先的前沿和開放權重模型之間進行選擇，以獲得最佳結果。客戶可以選擇在雲端或其本地環境中部署 Maestro 模型。

Anirudh Devgan - Cadence, CEO

(00:05:39)Maestro is gaining strong early customer traction, with several deployments currently underway. The momentum of our AI super agents continues to accelerate, showing compelling early results and engaging more than 100 active customers. We have added a new agent to ChipStack for RTL generation and early PPA optimization. This agent produced RTL that delivers an average of 24% area reduction and 18% power savings. This is compared to code generated purely by foundation models. Importantly, this demonstrates the power of combining our AI super agents with AI models and our underlying EDA engines to achieve results that foundation models alone cannot deliver.

(00:05:39)Maestro 正在獲得強勁的早期客戶吸引力，目前有幾個部署正在進行中。我們的 AI 超級代理的動能持續加速，顯示出引人注目的早期結果，並吸引了超過 100 名活躍客戶。我們已經為 ChipStack 新增了一個代理，用於 RTL 生成和早期 PPA 優化。這個代理生成的 RTL 提供了平均 24% 的面積減少和 18% 的功耗節省。這是與純粹由基礎模型生成的代碼相比較的。重要的是，這展示了將我們的 AI 超級代理與 AI 模型及我們的基礎 EDA 引擎結合的力量，以實現基礎模型無法單獨提供的結果。

Anirudh Devgan - Cadence, CEO

(00:06:41)Honda is utilizing the RTL generation agent and AI-powered automation to improve productivity from specification through RTL development. In Q3, MediaTek deployed ChipStack into its production digital verification flow, achieving a 50 to 70% improvement in verification coverage and a 1.4 to 2x reduction in engineering effort. Deployment is now scaling across multiple design teams. In analog and custom design, ViraStack saw a 4x increase in customer engagement, with deployments in various production designs at multiple customers.

(00:06:41)本田正在利用 RTL 生成代理和 AI 驅動的自動化來提高從規範到 RTL 開發的生產力。在第三季度，聯發科技將 ChipStack 部署到其生產數位驗證流程中，實現了 50% 到 70% 的驗證覆蓋率提升，以及 1.4 到 2 倍的工程努力減少。部署現在正在多個設計團隊中擴展。在類比和定製設計中，ViraStack 的客戶參與度增長了 4 倍，並在多個客戶的各種生產設計中進行了部署。

Anirudh Devgan - Cadence, CEO

(00:07:35)The combination of a agentic workflow, embedded skills, and our Virtuoso platform is unlocking significant new productivity opportunities in a traditionally difficult-to-automate domain. We are seeing strong customer interest in InnoStack, our digital AI super agent, as customers strive to improve PPA, accelerate design closure, and boost productivity. In Q3, the Renesas MCU team successfully deployed InnoStack across a wide range of engineering tasks, achieving a 34x productivity improvement, with plans to expand its use to layout automation.

(00:07:35)代理式（agentic）工作流程、嵌入式技能和我們的 Virtuoso 平台的結合正在解鎖顯著的新生產力機會在一個傳統上難以自動化的領域。我們看到客戶對 InnoStack 的強烈興趣，我們的數位 AI 超級代理，因為客戶努力改善 PPA，加速設計閉合，並提升生產力。在第三季度，瑞薩 MCU 團隊成功地在廣泛的工程任務中部署了 InnoStack，實現了 34 倍的生產力提升，並計劃將其使用擴展到佈局自動化。

Anirudh Devgan - Cadence, CEO

(00:08:26)Also in Q3, Samsung Foundry successfully used InnoStack to dramatically improve engineering efficiency and accelerate PPA exploration, reshaping its approach to technology migration and DTCO optimization. Socionext is deploying InnoStack in its production design environment and expects that combining it with our advanced physical design tools will deliver up to a 5x overall productivity gain while maintaining design quality. In PCB and 3D IC design, the early response to our AuraStack AI super agent has been overwhelmingly positive.

(00:08:26)同樣在第三季度，三星代工成功使用 InnoStack 顯著提高工程效率並加速 PPA 探索，重塑其技術遷移的方式和 DTCO 優化。Socionext 正在其生產設計環境中部署 InnoStack，並預期將其與我們的先進物理設計工具結合，將實現高達 5 倍的整體生產力增益，同時保持設計質量。在 PCB 和 3D IC 設計中，對我們的 AuraStack AI 超級代理的早期反應極為正面。

Anirudh Devgan - Cadence, CEO

(00:09:23)Nokia is deploying AuraStack on our Millennium supercomputer to advance electronic system design at scale. By unifying agentic AI, our physics-based solvers, and accelerated compute in a secure, on-premises platform, Nokia is achieving a 5x acceleration in key design and analysis workflows. Moving on to our businesses, we are delighted that all product groups delivered outstanding results. We achieved year-over-year revenue growth of 20% in core EDA, 24% in IP, and 30% in SDNA. Accelerating design activity continues to drive strong demand for our core EDA, SDA, and IP products.

(00:09:23)諾基亞正在我們的千年超級計算機上部署 AuraStack，以推進電子系統設計的規模。通過將代理 AI、我們基於物理的求解器和加速計算統一在一個安全的本地平台上，諾基亞在關鍵設計和分析工作流程中實現了 5 倍的加速。接下來談談我們的業務，我們很高興所有產品組都交出了優異的成績。我們在核心 EDA 中實現了 20% 的年增長，在 IP 中實現了 24%，在 SDNA 中實現了 30%。加速的設計活動持續推動我們的核心 EDA、SDA 和 IP 產品的強勁需求。

Anirudh Devgan - Cadence, CEO

(00:10:29)At the same time, we are seeing remarkable customer momentum for our agentic AI portfolio, which is also fueling increased demand for our underlying products. We are pleased that recurring revenue sustained its strong momentum, growing 18% year over year. I'm particularly proud of the significant, wide-ranging agreements we have signed with top customers. We furthered our long-standing partnership with a global marquee systems customer through a significant, broad expansion of our core EDA, hardware, and SDA solutions.

(00:10:29)同時，我們看到我們的代理 AI 產品組合的客戶動能顯著增強，這也促進了對我們基礎產品的需求增加。我們很高興經常性收入保持強勁動能，年增長 18%。我特別自豪於我們與頂級客戶簽署的重大、廣泛的協議。我們通過一項重大、廣泛的擴展進一步加強了與全球知名系統客戶的長期合作夥伴關係，擴展了我們的核心 EDA、硬體和 SDA 解決方案。

Anirudh Devgan - Cadence, CEO

(00:11:22)In Q3, Cadence and a top telco deepened their long-term strategic partnership with a broad expansion across our AI offerings, EDA software, hardware, and SDA solutions, establishing Cadence as their primary design partner. Core EDA revenue growth was driven by broad-based strength across the portfolio. Our digital and sign-off business continued to expand, achieving 16 new full-flow wins during the quarter. As agentic flows increasingly require full vertical workflows, there is growing demand for our synthesis and sign-off solutions.

(00:11:22)在第三季度，Cadence 和一家頂級電信商（a top telco） 深化了他們的長期戰略夥伴關係，通過我們的 AI 產品、EDA 軟體、硬體和 SDA 解決方案的廣泛擴展，確立了 Cadence 作為他們的主要設計夥伴。核心 EDA 收入增長是由整個產品組合的廣泛增強推動的。我們的數位和簽核業務持續擴展，在本季度實現了 16 個全流程的新勝利。隨著代理流程越來越需要完整的垂直工作流程，我們的綜合和簽核解決方案的需求也在增長。

Anirudh Devgan - Cadence, CEO

(00:12:16)Our sign-off product saw growing adoption, and our verification software products experienced strong growth in large digital and memory designs, with increasing usage at major hyperscalers and memory companies. As AI and HPC designs continue to grow in scale and complexity, hardware-assisted verification has become an increasingly strategic capability for our customers. This is driving sustained, unprecedented demand for our Palladium and Protium emulation platforms, which delivered yet another record quarter.

(00:12:16)我們的簽核產品的採用不斷增長，我們的驗證軟體產品在大型數位和記憶體設計中經歷了強勁增長，主要超大規模雲端業者和記憶體公司的使用量也在增加。隨著 AI 和 HPC 設計的規模和複雜性持續增長，硬體輔助驗證已成為我們客戶越來越重要的戰略能力。這推動了對我們的 Palladium 和 Protium 模擬平台持續、前所未有的需求，這些平台又交出了創紀錄的季度成績。

Anirudh Devgan - Cadence, CEO

(00:13:03)We added several new logos and saw meaningful expansion with marquee customers across a wide range of verticals. Our hardware cloud offering showed strong growth, with a leading semiconductor company adopting more hardware cloud services to accelerate new product development. In analog and custom design, SpectreFX continues to gain momentum, highlighted by major competitive displacement at a marquee infrastructure silicon provider. This includes multiple new two-nanometer projects adopting our platform.

(00:13:03)我們新增了幾個新客戶標誌，並在各個垂直領域與知名客戶實現了有意義的擴展。我們的硬體雲服務顯示出強勁增長，一家領先的半導體公司採用了更多的硬體雲服務以加速新產品開發。在類比和定製設計中，SpectreFX 繼續獲得動能，主要競爭對手在知名基礎設施矽供應商中被取代。這包括多個新的兩納米項目採用我們的平台。

Anirudh Devgan - Cadence, CEO

(00:13:54)Our IP business delivered another outstanding quarter, driven by deepening engagement with leading hyperscalers and semi-customers. We continue to benefit from strong AI-driven tailwinds, fueling accelerating AI infrastructure investments, growing adoption of custom silicon, and increasing demand for high-performance IP. Along with the transition to advanced nodes and chiplet-based architectures, these AI-driven trends are creating strong demand for a differentiated star IP portfolio across interface, memory, and foundation IP.

(00:13:54)我們的 IP 業務又交出了優異的季度成績，這是由於與領先的超大規模雲端業者和半導體客戶的深入合作。我們繼續受益於強勁的 AI 驅動的順風，推動加速的 AI 基礎設施投資，定製矽的採用增長，以及對高性能 IP 的需求增加。隨著向先進節點和基於芯片的架構的過渡，這些 AI 驅動的趨勢正在為跨界面、記憶體和基礎 IP 的差異化明星 IP 投資組合創造強勁需求。

Anirudh Devgan - Cadence, CEO

(00:14:40)We are well-positioned for sustained growth. As we continue to deliver superior power, performance, and area across our expanding advanced IP portfolio. A leading space technology company selected our IP portfolio for their custom chip program, paving the way for broader services and silicon engagement. Tensilica delivered a record revenue quarter. We also announced a joint development with Analog Devices across next-generation audio applications. Our system design and analysis business delivered another strong quarter, driven by growing adoption of our integrated solutions and the expansion of our electronics and physical systems intelligence portfolio.

(00:14:40)我們在持續增長方面處於良好位置。隨著我們持續提供卓越的功率、性能和範圍，擴展我們的先進 IP 產品組合。一家領先的太空科技公司選擇了我們的 IP 產品組合，用於其定製芯片計劃，為更廣泛的服務和矽晶體參與鋪平了道路。Tensilica 創造了創紀錄的收入季度。我們還宣佈與模擬設備公司在下一代音頻應用方面的聯合開發。我們的系統設計和分析業務交出了另一個強勁的季度，這得益於我們集成解決方案的日益採用以及我們電子和物理系統智能產品組合的擴展。

Anirudh Devgan - Cadence, CEO

(00:15:39)Allegro momentum continued. This included a competitive displacement at a leading electric vehicle customer. This demonstrates Cadence's integrated PCB design and multi-physics analysis platform as the primary solution. Our Millennium platform saw strong adoption, driven by deployment of our GPU-accelerated electronic and mechanical system suite. Along with our Reality Digital Twin solution. During the quarter, Switch, a leader in sustainable high-density AI infrastructure, selected our Millennium and Reality Digital Twin solutions to advance the design and simulation of its next-generation AI factories.

(00:15:39)Allegro 的勢頭持續增強。這包括在一家領先的電動車客戶中實現的競爭性替代。這證明瞭 Cadence 的集成 PCB 設計和多物理分析平台是主要解決方案。我們的 Millennium 平台得到了強勁的採用，這得益於我們 GPU 加速的電子和機械繫統套件的部署。以及我們的 Reality Digital Twin 解決方案。在本季度，Switch，一家可持續高密度 AI 基礎設施的領導者，選擇了我們的 Millennium 和 Reality Digital Twin 解決方案，以推進其下一代 AI 工廠的設計和模擬。

Anirudh Devgan - Cadence, CEO

(00:16:38)The MSC business is an integral part of our physical AI strategy, and its integration is progressing well. MSC Adams, our industry-leading multi-body dynamic solution, helped secure a competitive displacement at a leading power company, along with adoption of other Cadence products. In summary, Cadence delivered an exceptional Q3, and I am delighted with the accelerating momentum of our business. We are executing extremely well in a vibrant AI market, have our strongest competitive position ever, and are extending our leadership in AI.

(00:16:38)MSC 業務是我們物理 AI 策略的重要組成部分，其整合進展良好。MSC Adams，我們行業領先的多體動力學解決方案，幫助在一家領先的電力公司中實現了競爭性替代，並採用了其他 Cadence 產品。總結來說，Cadence 交出了卓越的第三季度，我對我們業務的加速勢頭感到高興。我們在充滿活力的 AI 市場中執行得非常出色，擁有有史以來最強的競爭地位，並在 AI 領域擴大我們的領導地位。

Anirudh Devgan - Cadence, CEO

(00:17:30)Our highly recurring and ratable business model, combined with growing customer R&D investment, is positioning us very well for continued strength in this dynamic AI environment. With that, I will turn it over to John to provide more details on our Q3 results and our updated 2026 outlook.

(00:17:30)我們高度重複和可計量的商業模式，加上客戶研發投資的增長，使我們在這個動態的 AI 環境中保持持續的強勁表現。接下來，我將把時間交給 John，讓他提供更多有關我們第三季度業績和更新的 2026 年展望的細節。

John Wall - Cadence, CFO

(00:17:57)Thanks, Anirudh, and good afternoon, everyone. Cadence delivered strong third-quarter results driven by robust demand across our businesses and growing momentum from AI-driven designs. Growth is accelerating and broadening. Q3 revenue grew 22% year over year, with strong double-digit growth across all product groups. We exceeded our guidance on every key financial metric and ended the quarter with a record backlog of $8.2 billion. Here are some of the financial highlights from the third quarter, starting with the P&L.

(00:17:57)謝謝你，Anirudh，大家下午好。Cadence 交出了強勁的第三季度業績，這得益於我們各業務的強勁需求和 AI 驅動設計的增長勢頭。增長正在加速並擴大。第三季度收入同比增長 22%，所有產品組別均實現了強勁的雙位數增長。我們在每一個關鍵財務指標上都超出了指導預測，並以 82 億美元的創紀錄訂單量結束了本季度。以下是第三季度的一些財務亮點，首先是損益表。

John Wall - Cadence, CFO

(00:18:32)Total revenue was $1,633,000,000. GAAP operating margin was 26.2%. Non-GAAP operating margin was 44.1%. GAAP EPS was $1.05. Non-GAAP EPS was $2.11. Next, turning to the balance sheet and cash flow, our cash balance was $1,452,000,000, while the principal value of debt outstanding was $2.6 billion. Operating cash flow was $424 million. DSOs were 66 days, and we used $400 million to repurchase Cadence shares during the quarter, reflecting our confidence in the business and our commitment to returning capital to shareholders.

(00:18:32)總收入為 16 億 3300 萬美元。GAAP 營業利潤率為 26.2%。非 GAAP 營業利潤率為 44.1%。GAAP 每股收益為 1.05 美元。非 GAAP 每股收益為 2.11 美元。接下來，轉向資產負債表和現金流，我們的現金餘額為 14 億 5200 萬美元，而未償還的債務本金為26 億美元。營運現金流為 4.24 億美元。應收賬款天數為 66 天，我們在本季度使用了 4 億美元回購 Cadence 股票，這反映了我們對業務的信心以及對股東回報資本的承諾。

John Wall - Cadence, CFO

(00:19:27)Underscoring that commitment, our board has authorized an additional $2.5 billion for our share repurchase program. Before I provide our updated outlook, I'd like to highlight that it assumes export control regulations currently in place remain substantially similar for the remainder of the year. For our updated outlook for 2026, we now expect revenue in the range of $6.387 billion to $6.417 billion. GAAP operating margin in the range of 28.2% to 29.2%. Non-GAAP operating margin in the range of 44.2% to 45.2%.

(00:19:27)強調這一承諾，我們的董事會已授權額外的 25 億美元用於我們的股票回購計劃。在我提供我們的更新展望之前，我想強調這假設目前的出口管制法規在今年剩餘時間內保持基本相似。對於我們 2026 年的更新展望，我們現在預計收入範圍為 63.87 億美元至 64.17 億美元。GAAP 營業利潤率範圍為 28.2% 至 29.2%。非 GAAP 營業利潤率範圍為 44.2% 至 45.2%。

John Wall - Cadence, CFO

(00:20:14)GAAP EPS in the range of $4.84 to $4.90. Non-GAAP EPS in the range of $8.37 to $8.43. Operating cash flow of approximately $2 billion, and we expect to use more than 50% of our free cash flow to repurchase Cadence shares in 2026. For Q4, we expect revenue in the range of $1,695,000,000 to $1,725,000,000. GAAP operating margin in the range of 30% to 31%, Non-GAAP operating margin in the range of 44% to 45%. GAAP EPS in the range of $1.23 to $1.29. And non-GAAP EPS in the range of $2.19 to $2.25. As usual, we published a CFO commentary document on our investor relations website, which includes our outlook for additional items, as well as further analysis and GAAP to non-GAAP reconciliations.

(00:20:14)GAAP 每股收益範圍為 4.84 美元至 4.90 美元。非 GAAP 每股收益範圍為 8.37 美元至 8.43 美元。營運現金流約為 20 億美元，我們預計在 2026 年將使用超過 50% 的自由現金流回購 Cadence 股票。對於第四季度，我們預計收入範圍為 16 億 9500 萬美元至 17 億 2500 萬美元。GAAP 營業利潤率範圍為 30% 至 31%，非 GAAP 營業利潤率範圍為 44% 至 45%。GAAP 每股收益範圍為 1.23 美元至 1.29 美元。非 GAAP 每股收益範圍為 2.19 美元至 2.25 美元。和往常一樣，我們在投資者關係網站上發布了 CFO 評論文件，其中包括我們對其他項目的展望，以及進一步的分析和 GAAP 與非 GAAP 的調整。

John Wall - Cadence, CFO

(00:21:35)Overall, I am pleased with our Q3 results, especially our recurring revenue growth of 18% year-over-year. At the midpoint of our updated 2026 outlook, we now expect revenue growth of 21%, an operating margin of 44.7%, and earnings per share of $8.40. As always, I would like to close by thanking our customers, partners, and employees for their continued support.

(00:21:35)總體而言，我對我們的第三季度業績感到滿意，特別是我們的經常性收入同比增長 18%。在我們更新的 2026 年展望的中點，我們現在預期收入增長 21%，營業利潤率為 44.7%，每股收益為 8.40 美元。一如既往，我想以感謝我們的客戶、合作夥伴和員工的持續支持作為結尾。

Q&A John Wall - Cadence, CFO

(00:22:02)With that, operator, we will now take questions.

(00:22:02)有鑑於此，操作員，我們現在將開始提問。

Jade - Conference Operator

(00:22:08)Thank you. At this time, I would like to remind everyone who wishes to ask a question to please press star, then the number one on your telephone keypad. To withdraw your question, press star one again. We ask that you pick up your handset when asking a question to ensure optimum sound quality. If you are muted locally, please remember to unmute your device. As a courtesy to all participants, please limit yourself to one question. We will pause briefly to compile the Q&A roster. Your first question comes from Jason Celino of KeyBanc Capital Markets.

(00:22:08)謝謝。此時，我想提醒所有希望提問的人，請按星號，然後按電話鍵盤上的數字一。若要撤回您的問題，再次按星號一。我們請您在提問時拿起話筒，以確保最佳的音質。如果您在本地靜音，請記得取消靜音您的設備。為了對所有參與者表示禮貌，請限制自己只提問一個問題。我們將暫停片刻以編制問答名單。您的第一個問題來自 KeyBanc Capital Markets 的 Jason Celino。

Jade - Conference Operator

(00:22:46)Please go ahead.

(00:22:46)請開始。

Jason Celino - KeyBanc Capital Markets, Analyst

(00:22:49)Thank you for taking my question. Anirudh, it’s great to see the Maestro announcement. I’m not an EDA engineer, nor an AI scientist. Could you please explain what this is a bit more clearly? It sounds like it’s part model, part router, but also maybe domain-specific models. Is Cadence developing these domain-specific models? Could you elaborate and explain how monetization of this new product will work? Thank you.

(00:22:49)謝謝您回答我的問題。Anirudh，很高興看到 Maestro 的公告。我不是 EDA 工程師，也不是 AI 科學家。您能否更清楚地解釋一下這是什麼？聽起來這部分是模型，部分是路由器，但也可能是特定領域的模型。Cadence 正在開發這些特定領域的模型嗎？您能否詳細說明一下這個新產品的貨幣化將如何運作？謝謝。

Anirudh Devgan - Cadence, CEO

(00:23:20)Yeah. Hi, Jason. Thank you for the question. There’s a lot in that release. Let me try to unpack it a bit. First of all, we are very pleased with the momentum of our Agentic AI stack and all four super agents. You can see there are now more than 100 engagements. As we have worked with customers over the last six to nine months, several trends have become clear. Okay. To me, we are launching Maestro as what I would call the next generation of agents. There are three major factors in this next generation of agents.

(00:23:20)是的。嗨，Jason。謝謝您的問題。那個公告裡有很多內容。讓我試著解釋一下。首先，我們對我們的 Agentic AI 堆疊和所有四個超級代理的勢頭感到非常高興。您可以看到現在有超過 100 個參與。在過去六到九個月中，我們與客戶合作時，幾個趨勢變得明顯。好的。對我來說，我們正在推出 Maestro，這是我所稱的下一代代理。這一代代理有三個主要因素。

Anirudh Devgan - Cadence, CEO

(00:24:00)One is what has happened over the last year, Of course, we have worked with all the frontier labs from the very beginning, whether they are doing chip design themselves or we are using their large language models. But over the last year, there has been the emergence of very high-quality open-weight models. I have always wanted to develop domain-specialized models, and now it is finally possible with open-weight models. It is actually very cost-effective to post-train these models to deliver domain intelligence, which is remarkable.

(00:24:00)一是過去一年發生的事情，當然，我們從一開始就與所有前沿實驗室合作，無論他們自己是否在進行芯片設計，或者我們是否在使用他們的大型語言模型。但在過去一年中，出現了非常高質量的開放權重模型。我一直想開發特定領域的專業模型，現在終於可以通過開放權重模型來實現。實際上，對這些模型進行後訓練以提供領域智能是非常具成本效益的，這是相當驚人的。

Anirudh Devgan - Cadence, CEO

(00:24:35)We are achieving up to 30% better quality by using our tools and data to post-train these models. Also, you don’t need just one model; it’s a portfolio of models. This ranges from small decision LLMs to small and medium LLMs, all the way up to frontier-class open-weight models post-trained on our data and intelligence. That is the first part of Maestro. It provides real customization for chip and system design and delivers much better results. It is quite remarkable how much these models can be improved by post-training.

(00:24:35)我們通過使用我們的工具和數據對這些模型進行後訓練，實現了高達 30% 的質量提升。此外，您不需要只有一個模型；這是一組模型。這範圍從小型決策 LLM 到小型和中型 LLM，一直到在我們的數據和智能上進行後訓練的前沿級開放權重模型。這是 Maestro 的第一部分。它為芯片和系統設計提供了真正的定製，並交付了更好的結果。這些模型通過後訓練可以得到多大的改善，實在是相當驚人。

Anirudh Devgan - Cadence, CEO

(00:25:13)Okay. The second part is that our agents perform a variety of tasks, so So, we don't need only one size or type of model to accomplish the task. The second part is model routing. Just like you wouldn't use an airplane to go to your local grocery store, right? Similarly, with agents, we have a variety of tasks ranging from simple ones to deep thinking and reasoning tasks. The model router efficiently identifies the task from our super agents and selects the appropriate model. This can easily provide up to a 5X improvement in token efficiency.

(00:25:13)好的。第二部分是我們的代理執行各種任務，所以所以，我們不需要只有一種大小或類型的模型來完成這項任務。第二部分是模型路由。就像你不會用飛機去當地的雜貨店，對吧？同樣地，對於代理，我們有各種各樣的任務，從簡單的任務到深度思考和推理的任務。模型路由器有效地從我們的超級代理中識別任務並選擇合適的模型。這可以輕鬆提供高達 5 倍的令牌效率提升。

Anirudh Devgan - Cadence, CEO

(00:25:53)This results in either significantly lower costs or much more work accomplished for the same cost. The third aspect, which I often discuss, is the reinforcement learning loop integrated with the underlying tools. We have a great example of this. For instance, our new RTL agent was publicly endorsed by a customer, Honda, which reported a 24% improvement.

(00:25:53)這導致顯著降低成本，或者以相同的成本完成更多的工作。第三個方面，我經常討論的是與基礎工具集成的強化學習循環。我們有一個很好的例子。例如，我們的新 RTL 代理得到了客戶本田的公開支持，報告顯示改善了 24%。

Anirudh Devgan - Cadence, CEO

(00:26:18)They achieved 24% area and 18% power reduction due to integration with the underlying tools and continuous self-improvement, whether in skills or models, giving us a unique advantage. Combining these three elements—domain-specialized models, model routing, and deep integration with our tools and data—creates a unique differentiation that no one else can match. This combination provides a unique differentiation that no other company can replicate. It also allows us to control and innovate across all three layers of the technology stack.

(00:26:18)他們因為與基礎工具的整合和持續的自我改進，無論是在技能還是模型上，實現了 24% 的面積和 18% 的功率減少，這給了我們獨特的優勢。結合這三個元素——領域專業模型、模型路由和與我們的工具及數據的深度整合——創造了獨特的差異化，沒有其他公司能夠匹敵。這種結合提供了其他公司無法複製的獨特差異化。它還使我們能夠在技術堆棧的所有三個層面上進行控制和創新。

Anirudh Devgan - Cadence, CEO

(00:26:57)I am genuinely very excited about this. I believe this approach will become standard across industries as large language models mature—domain-specialized intelligence, model routing, and integration with underlying engines.

(00:26:57)我對此感到非常興奮。我相信，隨著大型語言模型的成熟，這種方法將成為各行各業的標準——領域專業智能、模型路由和與基礎引擎的整合。

Jade - Conference Operator

(00:27:13)Your next question comes from Jim Schneider of Goldman Sachs. Please go ahead.

(00:27:13)您的下一個問題來自高盛的 Jim Schneider。請繼續。

Jim Schneider - Goldman Sachs, Analyst

(00:27:20)Good afternoon. Thank you for taking my question. Could you provide an update on the monetization of the Agentic AI solutions you are seeing? Last quarter, you mentioned recurring revenue growth of 20%. This quarter, it was 18%. Could you explain the difference in recurring revenue growth between these periods? Also, was there any acceleration in the Agentic AI portion of that recurring revenue? Thank you.

(00:27:20)下午好。感謝您回答我的問題。您能否提供有關您所看到的 Agentic AI 解決方案的貨幣化更新？上一季度，您提到經常性收入增長了 20%。本季度為 18%。您能否解釋這些期間經常性收入增長的差異？此外，Agentic AI 部分的經常性收入是否有加速？謝謝。

Anirudh Devgan - Cadence, CEO

(00:27:47)Thanks for the question. I'll start, and then John can add his comments. I'm very proud of the recurring revenue growth. This quarter, there were specific factors, such as changes in China revenue compared to Q3 last year, which are not present this year. Overall, recurring revenue is performing phenomenally well. To put it in perspective, last year recurring revenue grew by 11%. This quarter, it was 18%, and last quarter was also very strong. I expect recurring revenue to continue performing very well in 2026.

(00:27:47)感謝您的提問。我先開始，然後 John 可以補充他的意見。我對經常性收入的增長感到非常自豪。本季度，有一些具體因素，例如與去年第三季度相比，中國收入的變化，今年並不存在。總體而言，經常性收入表現非常出色。為了讓您瞭解，去年經常性收入增長了 11%。本季度為 18%，而上一季度也非常強勁。我預計經常性收入在 2026 年將繼續表現良好。

Anirudh Devgan - Cadence, CEO

(00:28:26)This growth results from our Agentic AI solutions, as well as Agentic AI driving increased usage of our base tools, which we anticipated. That's what we are observing. There is significant engagement with Agentic AI and substantial usage of our base tools. This combination provides strong growth for recurring revenue. Coupled with our ongoing strength in IP and hardware, this results in very strong overall performance.

(00:28:26)這一增長來自我們的 Agentic AI 解決方案，以及Agentic AI 推動了我們基礎工具的使用增加，這是我們預期的。這就是我們所觀察到的。與 Agentic AI 的互動非常密切，並且我們的基礎工具的使用量也很大。這種結合為經常性收入提供了強勁的增長。加上我們在 IP 和硬體方面的持續優勢，這導致了非常強勁的整體表現。

John Wall - Cadence, CFO

(00:28:58)I agree with that, Anirudh. It reflects strong demand across our software portfolio, healthy customer expansions, and continued strength in Core EDA. That's the right way to view the strong contribution of recurring revenue. We are very pleased with these results.

(00:28:58)我同意，Anirudh。這反映了我們軟體產品組合的強勁需求、健康的客戶擴展以及核心 EDA 的持續強勁。這是看待經常性收入強勁貢獻的正確方式。我們對這些結果非常滿意。

Jade - Conference Operator

(00:29:29)Your next question comes from Vivek Arya of Bank of America Securities. Please go ahead.

(00:29:29)您的下一個問題來自美國銀行證券的 Vivek Arya。請繼續。

Vivek Arya - BofA Securities, Analyst

(00:29:37)Thank you for taking my question. Anirudh, we recently heard three points from your competitor: that they can grow sales in the mid-teen percentage range next year and compound that to 30%, that they can achieve 50% margins, and that a royalty-based model could add nearly a billion dollars to their top line over the next several years. I realize this is not the call to provide a multi-year growth outlook, but I would appreciate your perspective on how you envision this growth and royalty component, specifically for Cadence, and how well Cadence is positioned to maintain or expand market share and leading operating margins over time.

(00:29:37)感謝您回答我的問題。Anirudh，我們最近聽到您的競爭對手提出三點：他們明年可以在中等百分比範圍內增長銷售，並將其複合到 30%，他們可以實現 50% 的利潤率，並且基於版稅的模型在未來幾年內可以為他們的營收增加近十億美元。我知道這不是提供多年的增長展望的電話，但我希望您能分享一下您對這種增長和版稅組件的看法，特別是對於 Cadence，以及 Cadence 在維持或擴大市場份額和隨著時間推移保持領先的營運利潤率方面的定位。

Vivek Arya - BofA Securities, Analyst

(00:30:21)Thank you.

(00:30:21)謝謝你。

Anirudh Devgan - Cadence, CEO

(00:30:23)Thank you for the question, Vivek. Overall, I am very pleased with our growth, not just in revenue but also in profitability and margin. We often discuss the rule of 40 metric, which combines revenue growth and margin. Based on our guidance, we expect about 21% revenue growth and slightly less than 45% margin, resulting in a figure north of 65% in 2026. That's an important point to highlight. We still have one more quarter to complete. So we'll see how it unfolds; this quarter is important for us. The fourth quarter is always significant.

(00:30:23)謝謝你的提問，Vivek。總體而言，我對我們的增長感到非常滿意，不僅在收入上，還在盈利能力和利潤率上。我們經常討論 40 法則指標，這結合了收入增長和利潤率。根據我們的指導，我們預計 2026 年收入增長約 21%，利潤率略低於 45%，最終將達到 65% 以上。這是一個重要的重點。我們還有一個季度要完成。所以我們將看看情況如何發展；這個季度對我們來說很重要。第四季度總是很重要。

Anirudh Devgan - Cadence, CEO

(00:31:09)We'll see how it finishes, and as usual, we will update you in February regarding the next year. Overall, the environment remains very positive for us. As you know, the AI build-out continues. There is substantial design activity across all aspects of AI. This includes traditional companies, analog mixed-signal companies, digital companies, memory companies, and system companies. The level of design activity is the strongest I have seen. We are gaining share in almost all segments of our business, whether in IP, hardware, or EDA, including Agentic AI.

(00:31:09)我們將看看它的結束情況，像往常一樣，我們會在二月份更新你們關於明年的情況。總體而言，環境對我們來說仍然非常積極。如你所知，人工智慧的建設仍在繼續。在人工智慧的各個方面都有大量的設計活動。這包括傳統公司、類比混合信號公司、數字公司、記憶體公司和系統公司。我所見的設計活動水平是最強的。我們在幾乎所有業務領域都在獲得市場份額，無論是在 IP、硬體還是 EDA，包括 Agentic AI。

Anirudh Devgan - Cadence, CEO

(00:31:51)Our product portfolio is the strongest it has ever been. Additionally, we have a huge opportunity with Agentic AI. We are redefining how Agentic AI will be applied across industries with our own domain-specialized models, while maintaining flexibility to work with all frontier models. Considering all these factors, I believe Cadence is in a very strong position. Regarding your specific question on royalty, we are always exploring different business models. This includes various approaches to licensing and revenue generation.

(00:31:51)我們的產品組合是有史以來最強的。此外，我們在 Agentic AI 方面有巨大的機會。我們正在重新定義 Agentic AI 如何在各行各業中應用，並使用我們自己的領域專業模型，同時保持與所有前沿模型合作的靈活性。考慮到所有這些因素，我相信 Cadence 處於非常強大的位置。關於你具體提到的版稅問題，我們一直在探索不同的商業模式。這包括各種授權和收入生成的方法。

Anirudh Devgan - Cadence, CEO

(00:32:28)In IP, we have had royalties for some time in Tensilica, and now we also have royalties in our design IP business. However, royalty is just one aspect of our strategy. We continuously seek ways to deliver value to our customers and, in turn, receive appropriate value ourselves.

(00:32:28)在 IP 方面，我們在 Tensilica 中已經有了一段時間的版稅，現在我們的設計 IP 業務中也有版稅。然而，版稅只是我們策略的一個方面。我們不斷尋求為客戶提供價值的方法，並反過來獲得適當的價值。

Jade - Conference Operator

(00:32:51)Your next question comes from Siti Panigrahi of Mizuho. Please go ahead.

(00:32:51)你的下一個問題來自 Mizuho 的 Siti Panigrahi。請繼續。

Siti Panigrahi - Mizuho, Analyst

(00:32:59)Thank you. Congratulations on a strong quarter. I have a question regarding hardware. I noticed that inventory increased again this quarter. Given that hardware had a strong quarter, how sustainable is hardware growth heading into 2027? Any insights would be helpful. Also, a quick clarification for John. There was a spike in G&A expense in Q3. Was there any one-time item contributing to that? Thank you.

(00:32:59)謝謝你。恭喜你們取得了強勁的季度表現。我有一個關於硬體的問題。我注意到這個季度庫存再次增加。鑑於硬體表現強勁，硬體增長在 2027 年之前的可持續性如何？任何見解都將是有幫助的。此外，對 John 的快速澄清。第三季度的 G&A 費用出現了激增。是否有任何一次性項目導致了這一點？謝謝你。

John Wall - Cadence, CFO

(00:33:33)Sure, Siti. Great. I'll address the hardware question first. Hardware continues to perform exceptionally well. We delivered another record quarter, supported by strong demand from AI and HPC customers. We are also seeing growing interest in cloud-based hardware access. Hardware-assisted verification is increasingly important as customers design systems of unprecedented complexity and require capacity to verify designs before committing to silicon. We have a strong pipeline entering Q4 and continue to expect another record year for hardware.

(00:33:33)當然，Siti。很好。我將首先回答硬體問題。硬體的表現持續非常出色。我們交出了另一個創紀錄的季度，得益於來自 AI 和 HPC 客戶的強勁需求。我們也看到對基於雲的硬體訪問的興趣日益增長。隨著客戶設計前所未有的複雜系統，硬體輔助驗證變得越來越重要，客戶需要在承諾使用矽之前驗證設計的能力。我們在進入第四季度時擁有強大的管道，並繼續預期硬體將迎來另一個創紀錄的年份。

John Wall - Cadence, CFO

(00:34:04)As always, hardware results can be lumpy quarter to quarter, depending on product deliveries and customer installation schedules. However, the underlying demand environment remains very strong and healthy, and we are encouraged by the pipeline of customer engagement. Regarding inventory, the increase in hardware inventories supports customer orders, as we have a significant backlog we are working through. There are no capacity or supply chain constraints in addressing this backlog. We are building inventory to meet those order numbers.

(00:34:04)一如既往，硬體的結果可能會因產品交付和客戶安裝時間表而有所波動。然而，基本需求環境仍然非常強勁和健康，我們對客戶參與的管道感到鼓舞。關於庫存，硬體庫存的增加支持客戶訂單，因為我們有一個顯著的積壓正在處理。在解決這一積壓方面沒有產能或供應鏈的限制。我們正在建立庫存以滿足這些訂單數量。

John Wall - Cadence, CFO

(00:34:36)You also asked about G&A expenses. As part of the quarterly movement, this quarter reflects the timing of annual charitable contributions, which occurred in Q3 this year compared to Q4 last year.

(00:34:36)你還問到了 G&A 費用。作為季度變動的一部分，本季度反映了年度慈善捐款的時機，這些捐款在今年的第三季度進行，而不是去年的第四季度。

Anirudh Devgan - Cadence, CEO

(00:34:49)That timing difference slightly impacts the underlying G&A run rate, causing a minor effect in G&A for Q3 due to those charitable contributions. Regarding hardware, Palladium and Protium are performing exceptionally well, not only for AI but also for other chip design applications. We are collaborating with all the major AI companies, and hardware has become an indispensable tool for design. Hardware is now almost indispensable for designing these systems. One question is whether this trend will continue.

(00:34:49)這種時機差異稍微影響了基礎的管理和行政費用（G&A）運行率，因為這些慈善捐款對第三季度的 G&A 造成了輕微影響。關於硬體，Palladium 和 Protium 的表現非常出色，不僅在人工智慧領域，也在其他晶片設計應用中。我們正在與所有主要的人工智慧公司合作，硬體已成為設計中不可或缺的工具。硬體現在幾乎是設計這些系統不可或缺的。一個問題是這一趨勢是否會持續下去。

Anirudh Devgan - Cadence, CEO

(00:35:28)We'll see, but there is one mega trend to consider. As you probably know, hardware is typically sold or consumed based on design size. So, the larger the design size, the more hardware is required. Recently, I considered what will happen to design size in the future. Design sizes are already large, but what will happen going forward? For example, I recently attended IMEC. I gave a talk at IMEC, a European research consortium, where TSMC also presented in the same session. TSMC showed their process roadmap, projecting that chip size will increase by 48 times over the next five years.

(00:35:28)我們拭目以待，但有一個重大趨勢需要考慮。正如你們可能知道的，硬體通常是根據設計大小來銷售或消耗的。因此，設計越大，所需的硬體就越多。最近，我考慮了未來設計大小會發生什麼變化。設計大小已經很大，但未來會發生什麼呢？例如，我最近參加了 IMEC。我在 IMEC 上發表了演講，這是一個歐洲研究聯盟，台積電也在同一場次中發表了報告。台積電展示了他們的製程路線圖，預測晶片大小在未來五年內將增加 48 倍。

Anirudh Devgan - Cadence, CEO

(00:36:17)Typically, such growth would be 5 to 10 times, but this 48x increase is due to regular technology scaling plus 3D IC packaging for scaled integration. This will further drive demand for hardware, die-to-die IP, and related technologies. Our customers are now verifying not just one chip or chiplet, but multiple chiplets working together, with both chiplet size and quantity increasing. This trend of hardware-based design and verification, combined with growth in design size, is expected to continue for some time.

(00:36:17)通常，這樣的增長會是 5 到 10 倍，但這 48 倍的增長是由於常規技術的縮放加上 3D IC 封裝的縮放整合。這將進一步推動對硬體、晶片對晶片的 IP 和相關技術的需求。我們的客戶現在不僅在驗證一個晶片或晶片組，而是在驗證多個晶片組的協同工作，晶片組的大小和數量都在增加。這種基於硬體的設計和驗證趨勢，加上設計大小的增長，預計將持續一段時間。

Jade - Conference Operator

(00:37:07)Your next question comes from Joe Quatrochi of Wells Fargo. Please go ahead.

(00:37:07)你的下一個問題來自於 Wells Fargo 的 Joe Quatrochi。請繼續。

Joe Quatrochi - Wells Fargo, Analyst

(00:37:15)Thank you for taking my question. I’m curious about the change in the 2026 outlook, particularly regarding operating profit. When considering the incremental margin on the revenue upside, it seems quite significant compared to your target of over 50%. Could you help us understand what is driving this?

(00:37:15)感謝你回答我的問題。我對 2026 年的展望變化感到好奇，特別是關於營業利潤的部分。考慮到收入上升的增量利潤，這似乎相對於你們超過 50% 的目標來說相當顯著。你能幫助我們理解這背後的驅動因素嗎？

John Wall - Cadence, CFO

(00:37:37)Sure, happy to explain. We remain very pleased with the underlying profitability of the business. Despite integrating Hexagon's D&E business and making targeted investments in strategic growth areas, we now expect approximately a 44.7% non-GAAP operating margin for 2026, slightly above last year. Revenue growth this year over last is about $1.1 billion. Non-GAAP operating profit is growing by about $500 million. Considering Hexagon D&E is part of this, and it is not contributing much operating profit this year, we initially expected Hexagon D&E to contribute about $160 million [待確認：營收或營業利益] for the year.

(00:37:37)當然，很高興能解釋。我們對業務的基礎盈利能力感到非常滿意。儘管整合了 Hexagon 的 D&E 業務並在戰略增長領域進行了針對性投資，我們現在預計 2026 年的非 GAAP 營業利潤率約為 44.7%，略高於去年。今年的收入增長與去年相比約為 11 億美元。非 GAAP 營業利潤增長約為 5 億美元。考慮到 Hexagon D&E 是其中的一部分，而它今年對營業利潤的貢獻不大，我們最初預計 Hexagon D&E 今年的貢獻約為 1.6 億美元[待確認：營收或營業利益]。它的表現略好於預期。

John Wall - Cadence, CFO

(00:38:18)It is performing slightly better than that. But it is generally on target. Excluding that, we would be in our typical range: mid-teens revenue growth, high teens EPS growth, and an incremental margin around 54-55%. The additional D&E business is impacting incremental margin this year. However, we expect it to benefit us next year as integration progresses well. We are taking steps now to improve performance next year, similar to what we did with BETA over the past couple of years.

(00:38:18)但總體上仍在目標範圍內。排除這部分，我們將在典型範圍內：中等十幾的收入增長，高十幾的每股收益增長，以及約 54-55% 的增量利潤率。今年額外的 D&E 業務對增量利潤率產生了影響。然而，我們預計隨著整合的順利進行，明年將對我們有利。我們現在正在採取措施，以改善明年的表現，類似於我們在過去幾年對 BETA 所做的。你的下一個問題來自 Needham & Company 的 Charles Shi。

Jade - Conference Operator

(00:38:58)Your next question comes from Charles Shi of Needham & Company. Please go ahead.

(00:38:58)請繼續。感謝你回答我的問題。

Charles Shi - Needham & Company, Analyst

(00:39:06)Thank you for taking my question. I’m glad to hear about the Maestro model portfolio announced today. In light of your competitor announcing GPT-Synopsys, can you share your thoughts on collaborating with Frontier Labs for a specialized model? Given that Maestro, as you described, appears to be a collection of open models post-trained by Cadence, and now your customers have choices between your model portfolio and GPT-Synopsys, what do you tell customers when they have both options, and why should they choose Cadence?

(00:39:06)我很高興聽到今天宣佈的 Maestro 模型組合。鑒於你的競爭對手宣佈了 GPT-Synopsys，你能分享一下與 Frontier Labs 合作開發專門模型的想法嗎？鑒於 Maestro，如你所描述的，似乎是 Cadence 後期訓練的開放模型的集合，現在你的客戶在你的模型組合和 GPT-Synopsys 之間有選擇，當他們有這兩個選擇時，你會告訴客戶什麼，為什麼他們應該選擇 Cadence？謝謝。

Charles Shi - Needham & Company, Analyst

(00:39:56)Thank you.

(00:39:56)感謝你的問題，Charles。

Anirudh Devgan - Cadence, CEO

(00:39:58)Thanks for the question, Charles. First, I want to point out that we have worked with OpenAI from the very beginning, collaborating on chip design and various projects. And the same applies to Anthropic and Gemini, the three leading frontier models. We consistently discuss various collaboration opportunities with them. Now, regarding what we want to ensure as we develop Maestro, The key point is that Maestro is not limited to our own models; it maintains maximum flexibility to use the most suitable model for each task.

(00:39:58)首先，我想指出的是我們從一開始就與 OpenAI 合作，協作進行晶片設計和各種項目。同樣的情況也適用於 Anthropic 和 Gemini，這三個領先的前沿模型。我們持續與他們討論各種合作機會。現在，關於我們在開發 Maestro 時想要確保的事項，關鍵點是 Maestro 並不僅限於我們自己的模型；它保持最大靈活性，以便為每個任務使用最合適的模型。

Anirudh Devgan - Cadence, CEO

(00:40:41)For example, if a certain frontier model performs better for a specific portion of the task, we will, of course, utilize it to deliver the best performance to our customers. However, the core of our solution is that we want to control all three layers of the stack and manage the interaction with our customers. We aim to be the primary interface for our customers, which aligns with their preferences. In return, our agents will provide significant value by using the appropriate large language model (LLM) for each task.

(00:40:41)例如，如果某個前沿模型在特定任務部分的表現更佳，我們當然會利用它來為我們的客戶提供最佳表現。然而，我們解決方案的核心是我們希望控制整個堆疊的三個層級，並管理與客戶的互動。我們的目標是成為客戶的主要介面，這與他們的偏好相符。作為回報，我們的代理將通過為每個任務使用適當的大型語言模型（LLM）來提供顯著的價值。

Anirudh Devgan - Cadence, CEO

(00:41:16)Crucially, as we become the primary interface, we guarantee the confidentiality of both our data and our customers' data. Additionally, Maestro will allow the following capabilities: If a customer wishes to further fine-tune Maestro, which is a common request, we will support that. If customers prefer to deploy it on-premises, we will accommodate that. Similarly, if customers want to host it on the cloud, we will enable that option. Okay. This approach provides maximum flexibility and performance by integrating all three layers.

(00:41:16)關鍵是，當我們成為主要介面時，我們保證我們的數據和客戶的數據的機密性。此外，Maestro 將允許以下功能：如果客戶希望進一步微調 Maestro，這是一個常見的請求，我們將支持這一點。如果客戶更喜歡在本地部署，我們將滿足這一需求。同樣，如果客戶希望在雲端託管，我們將啟用這一選項。好的。這種方法通過整合所有三個層級提供了最大的靈活性和性能。

Anirudh Devgan - Cadence, CEO

(00:41:47)Only Cadence can offer this level of integration. This will undoubtedly deliver the best results for our customers.

(00:41:47)只有 Cadence 能提供這種級別的整合。這無疑會為我們的客戶帶來最佳結果。

(00:41:53)Okay. Your next question comes from Harlan Sur of JPMorgan. Please go ahead.

(00:41:53)好的。您的下一個問題來自 JPMorgan 的 Harlan Sur。請繼續。

Harlan Sur - JPMorgan, Analyst

(00:42:05)Yes, good afternoon. Thank you for taking my question. Anirudh, last quarter you mentioned a pickup in the more traditional analog, power, and microcontroller segments of the market, targeting industrial, automotive, consumer, and IoT sectors. These are the positive cyclical dynamics we're seeing now as excess customer inventories have bottomed out. in these end markets. Are you continuing to observe positive demand trends in this segment? These same companies are now very focused on embedding next-generation physical AI capabilities into their chip designs.

(00:42:05)是的，下午好。感謝您回答我的問題。Anirudh，上個季度您提到傳統的類比、電源和微控制器市場出現回升，目標是工業、汽車、消費者和物聯網領域。這些是我們現在看到的正面循環動態，因為客戶的過剩庫存已經觸底。在這些最終市場中。您是否繼續觀察到這一領域的正面需求趨勢？這些公司現在非常專注於將下一代物理 AI 能力嵌入他們的芯片設計中。

Harlan Sur - JPMorgan, Analyst

(00:42:37)For example, your MCU customers are adding neural processing unit (NPU) capabilities to their chip designs. Your analog customers are designing more sensor-related features into their chips, including more A-to-D and D-to-A converters and on-chip DSP processing. Is this translating into a stronger book of business for the Cadence team?

(00:42:37)例如，您的 MCU 客戶正在添加神經處理單元（NPU）功能到他們的芯片設計中。您的類比客戶正在將更多傳感器相關功能設計到他們的芯片中，包括更多的 A 到 D 和 D 到 A 轉換器以及片上 DSP 處理。這是否轉化為 Cadence 團隊更強的業務訂單？

Anirudh Devgan - Cadence, CEO

(00:42:59)Yes, Harlan. That's an excellent point. I believe the momentum is very strong in analog and mixed-signal, as reflected in their results. This is driven by physical AI and the overall market recovery. Additionally, memory, where we collaborate with all the major companies, has shown remarkable improvement compared to a year or two ago. What this demonstrates is that we are very strong in big-signal companies. We have the most comprehensive portfolio, covering both analog and digital. These designs, as you mentioned, are increasingly mixed-signal.

(00:42:59)是的，Harlan。這是一個很好的觀點。我相信在類比和混合信號領域的動能非常強勁，這在他們的業績中得到了體現。這是由物理 AI 和整體市場復甦推動的。此外，記憶體領域，我們與所有主要公司合作，與一兩年前相比顯示出顯著改善。這表明我們在大信號公司中非常強大。我們擁有最全面的產品組合，涵蓋類比和數字。正如您所提到的，這些設計越來越多地是混合信號。

Anirudh Devgan - Cadence, CEO

(00:43:35)This trend plays to Cadence's strengths. Also, I am very pleased with the momentum in ViraStack. Analog customers, whether mixed-signal companies or large AI firms, have substantial analog teams. You might not realize that. Some of these major AI companies design their own IPs and maintain large analog teams. There is massive demand for analog productivity, which we could not fully address before. We always aimed to, but with Agentic AI now, we can significantly enhance analog productivity. This is evident in some of the customer codes and the volume of engagements we have with ViraStack.

(00:43:35)這一趨勢符合 Cadence 的優勢。此外，我對 ViraStack 的動能感到非常滿意。類比客戶，無論是混合信號公司還是大型 AI 公司，都擁有龐大的類比團隊。您可能沒有意識到這一點。一些主要的 AI 公司設計自己的 IP 並維持大型類比團隊。對於類比生產力的需求巨大，而我們之前無法完全滿足。我們一直希望這樣做，但現在有了 Agentic AI，我們可以顯著提升類比生產力。這在一些客戶代碼和我們與 ViraStack 的參與量中顯而易見。

Anirudh Devgan - Cadence, CEO

(00:44:22)Overall, I am very pleased not only with the analog mixed-signal environment but also with the new capabilities we are launching with ViraStack.

(00:44:22)總體而言，我對類比混合信號環境以及我們與 ViraStack 推出的新能力感到非常滿意。

Jade - Conference Operator

(00:44:34)Your next question comes from Joe Vruwink of Baird. One moment, please go ahead. Joe Vruwink, please remember to unmute your device. Your line is now open. Moving on, your next question comes from Lee Simpson of Morgan Stanley. Please go ahead.

(00:44:34)您的下一個問題來自 Baird 的 Joe Vruwink。一分鐘，請繼續。Joe Vruwink，請記得解除靜音。您的線路現在已開通。接下來，您的下一個問題來自摩根士丹利的 Lee Simpson。請繼續。

Lee Simpson - Morgan Stanley, Analyst

(00:45:15)Great. Thanks for fitting me in here. Anirudh, it seems that you and your peer are pursuing very distinct approaches to monetizing agentic AI in chip design. I thought you answered one of the previous questions quite well. It really comes through that you want to maintain control over the modeling. Your deployment flexibility is evident. So, you believe that adoption, as you see it, versus a frontier model-centric partnership, will gain strong traction. But is there something about the performance or delivery costs that led you to choose this strategic path?

(00:45:15)很好。感謝您讓我參與這次會議。Anirudh，似乎您和您的同行在芯片設計中對於代理 AI 的貨幣化採取了非常不同的方法。我認為您對之前的問題回答得相當好。您想要保持對建模的控制，這一點非常明顯。您的部署靈活性顯而易見。所以，您認為採用，正如您所見，與以前沿模型為中心的合作夥伴關係相比，將會獲得強勁的吸引力。但是，有沒有什麼關於性能或交付成本的因素使您選擇了這條戰略路徑？

Lee Simpson - Morgan Stanley, Analyst

(00:45:58)And maybe just a follow-up question. You mentioned that a leading space company is adopting some of your IP features. Is there anything specific you can highlight there? Are they taking typical IP blocks, or does it include broader IP services? Thanks.

(00:45:58)也許再問一個跟進問題。您提到一家領先的太空公司正在採用您的一些 IP 特徵。您能否具體指出一些內容？他們是採用典型的 IP 模塊，還是包括更廣泛的 IP 服務？謝謝。

Anirudh Devgan - Cadence, CEO

(00:46:16)Let me answer the second part first, then come back to the first. Generally, as we've mentioned, 45% of our business now comes from system companies—the non-traditional versus traditional semiconductor firms. These are new customers, and it's encouraging that it's not just the large AI hyperscalers designing chips, but also many other system companies, like the space company we highlighted, and even firms in finance designing their own chips. As chip design becomes easier, more companies will engage in it.

(00:46:16)讓我先回答第二部分，然後再回到第一部分。一般來說，正如我們提到的，45% 的業務現在來自系統公司——非傳統與傳統半導體公司。這些是新客戶，令人鼓舞的是，不僅僅是大型 AI 超大規模雲端業者在設計芯片，還有許多其他系統公司，例如我們提到的太空公司，甚至金融公司也在設計自己的芯片。隨著芯片設計變得更容易，更多公司將會參與其中。

Anirudh Devgan - Cadence, CEO

(00:46:59)When these new companies design chips, we naturally have IP opportunities, hardware sales, and design services, as they are just starting out. We have one of the best design services capabilities, being experts in using our own products. This is an area to watch, and I'm encouraged by more system companies entering chip design. Now, returning to the AI solution, we want to maintain control and provide flexibility to our customers. This is the right perspective. There is significant value in the model itself.

(00:46:59)當這些新公司設計芯片時，我們自然會有 IP 機會、硬體銷售和設計服務，因為他們剛剛起步。我們擁有最佳的設計服務能力之一，專精於使用我們自己的產品。這是一個值得關注的領域，我對更多系統公司進入芯片設計感到鼓舞。現在，回到AI 解決方案，我們希望保持控制並為客戶提供靈活性。這是正確的觀點。模型本身具有顯著的價值。

Anirudh Devgan - Cadence, CEO

(00:47:41)It is actually quite cost-effective to post-train a frontier-class model. This was not feasible two years ago. But now, the cost is manageable, and you achieve remarkable results, as I mentioned. The incremental cost to post-train a frontier model is very low. Of course, having this capability alongside a commercially available frontier model is important. It's about choosing the best solution between the two. In our own agentic development, we have our IP team and many people using Agentic AI and all related solutions.

(00:47:41)實際上，對前沿級模型進行後訓練是相當具成本效益的。這在兩年前是不可行的。但現在，成本是可控的，您可以獲得顯著的結果，正如我提到的。對前沿模型進行後訓練的增量成本非常低。當然，擁有這種能力與商業可用的前沿模型並存是重要的。這是關於在兩者之間選擇最佳解決方案。在我們自己的代理開發中，我們有 IP 團隊和許多人使用代理 AI 及所有相關解決方案。

Anirudh Devgan - Cadence, CEO

(00:48:20)The cost benefits apply to us as well. So, although there is some incremental expense to train these models, there are also savings in our AI deployment. Overall, this should not impact Cadence's financial model. It only strengthens it by adding new revenue capabilities, with margin progression continuing as planned. Maestro provides flexibility and customization ability. It offers control and cost savings. This represents the best of all worlds.

(00:48:20)成本效益對我們來說也是適用的。所以，儘管訓練這些模型會有一些增量費用，但在我們的 AI 部署中也有節省。總體而言，這不應影響 Cadence 的財務模型。它只會通過增加新的收入能力來加強它，利潤率的進展將繼續按計劃進行。Maestro 提供靈活性和定製能力。它提供控制和成本節省。這代表了各種優勢的最佳結合。

John Wall - Cadence, CFO

(00:48:59)Yes, Lee, if I may add, Anirudh makes a very important point about flexibility. We want customers to have choice and control over the models they use, while benefiting from Cadence's domain expertise and deep integration with our design tools. Our strategy remains focused on delivering high-value engineering software rather than building a large general-purpose AI infrastructure business. We are focused on improving the efficiency and value of agentic workflows while maintaining attractive software economics as customer adoption scales.

(00:48:59)是的，Lee，如果我可以補充，Anirudh 對靈活性提出了一個非常重要的觀點。我們希望客戶能夠選擇和控制他們使用的模型，同時受益於 Cadence 的領域專業知識和與我們設計工具的深度整合。我們的策略仍然專注於提供高價值的工程軟體，而不是建立一個大型通用 AI 基礎設施業務。我們專注於提高代理工作流程的效率和價值，同時在客戶採用擴大時保持吸引人的軟體經濟性。

John Wall - Cadence, CFO

(00:49:28)But we believe that flexibility is a key strength of our model. Our model allows customers to use their own LLM infrastructure and bear the associated compute costs if they choose to. CAITs [待確認] monetize the underlying design agents and our agentic orchestration capabilities. The economics of specific deployment options can vary, but the principle remains that greater customer usage should generate greater value for CAITs. And I believe that is where we truly differentiate ourselves.

(00:49:28)但我們相信，靈活性是我們模型的一個關鍵優勢。我們的模型允許客戶使用他們自己的 LLM 基礎設施，並承擔相關的計算成本，如果他們選擇這樣做。CAITs[待確認]獲利於底層設計代理和我們的代理協調能力。具體部署選項的經濟學可能會有所不同，但原則仍然是，客戶的使用越多，CAITs 應該產生的價值就越大。我相信這就是我們真正區別於其他公司的地方。

Jade - Conference Operator

(00:49:54)Your next question comes from Joe Vruwink of Baird. Please go ahead.

(00:49:54)您的下一個問題來自 Baird 的 Joe Vruwink。請繼續。

Joe Vruwink - Baird, Analyst

(00:50:03)Hi. Can you hear me?

(00:50:03)嗨。你能聽到我嗎？

Anirudh Devgan - Cadence, CEO

(00:50:05)Yes. Yes, Joe. Please go ahead.

(00:50:05)是的。是的，Joe。請繼續。

Joe Vruwink - Baird, Analyst

(00:50:07)Great. I wanted to revisit InnoStack and ViraStack. I think it's fair to say that the analog and backend areas haven't yet experienced the AI proliferation that frontend design has. Are you seeing this beginning to change? Are specialized models and post-training helping to better address these domains? And if this is starting to happen, does it imply a correspondingly higher revenue potential for Cadence compared to what we've seen so far?

(00:50:07)很好。我想重新討論 InnoStack 和 ViraStack。我認為可以公平地說，類比和後端領域尚未經歷前端設計的 AI 擴散。你們看到這開始改變了嗎？專門模型和後訓練是否有助於更好地解決這些領域？如果這開始發生，這是否意味著 Cadence 的收入潛力相對於我們目前所見會更高？

Anirudh Devgan - Cadence, CEO

(00:50:44)Hi, Joe. First of all, we started with ChipStack, and then we launched three other major super agents: ViraStack for analog, InnoStack for backend, and AuraStack, the latest, for package and PCB. The demand for these agents has been phenomenal. Demand is truly exceptional. Especially since ChipStack is performing well, and these other agents are even more integrated with our base tools. When we work with customers to obtain public endorsements, it typically happens with only about one in ten customers, as not everyone wants to disclose their activities.

(00:50:44)嗨，Joe。首先，我們從 ChipStack 開始，然後推出了三個其他主要超級代理：針對類比的 ViraStack，針對後端的 InnoStack，以及最新的 AuraStack，針對封裝和 PCB。這些代理的需求非常驚人。需求確實非常卓越。特別是因為 ChipStack 表現良好，而這些其他代理與我們的基礎工具更為整合。當我們與客戶合作以獲得公開背書時，通常只有大約十分之一的客戶會這樣做，因為並不是每個人都想披露他們的活動。

Anirudh Devgan - Cadence, CEO

(00:51:18)However, this quarter we have secured around ten endorsements. This indicates the broad deployment of all these agents. For example, Renesas reported a 34X productivity improvement with InnoStack, and Samsung Foundry has other use cases beyond physical design. One significant use case is design technology co-optimization with the foundry. InnoStack is a critical agent for that purpose. Socionext reported a 5X productivity improvement. Rapidus has reported a 2X improvement. So, multiple-fold productivity gains are occurring with InnoStack.

(00:51:18)然而，本季度我們已獲得約十個背書。這表明所有這些代理的廣泛部署。例如，Renesas 報告使用 InnoStack 提高了 34 倍的生產力，而三星代工廠在物理設計之外還有其他用例。一個重要的用例是與代工廠的設計技術共同優化。InnoStack 是實現這一目的的關鍵代理。Socionext 報告提高了 5 倍的生產力。Rapidus 報告提高了 2 倍的生產力。因此，使用 InnoStack 的生產力增長是多重的。

Anirudh Devgan - Cadence, CEO

(00:52:01)ViraStack demand remains off the charts. AuraStack is the latest area where innovation had not yet occurred. Nokia demonstrated a 5X productivity improvement with AuraStack. ChipStack has been deployed for about a year now, while the others have been active for six to nine months. Deployment of these newer agents is accelerating, and they are even more tightly integrated with our base tools. Maestro adds further differentiation on top of these capabilities. I feel very confident about this progress. We continue to invest heavily in R&D for our traditional tools in the middle layer.

(00:52:01)ViraStack 的需求仍然高得驚人。AuraStack 是創新尚未發生的最新領域。Nokia 使用 AuraStack 展示了 5 倍的生產力提升。ChipStack 已經部署約一年，而其他代理則活躍了六到九個月。這些新代理的部署正在加速，並且它們與我們的基礎工具更加緊密整合。Maestro 在這些能力之上進一步增強了差異化。我對這一進展感到非常有信心。我們繼續在中層的傳統工具上進行大量投資。

Anirudh Devgan - Cadence, CEO

(00:52:37)But now, we are making significant investments and differentiation in the top layer. We are defining what will be the agentic flow for the future across all industries. As we establish the agentic layer at the top, and as it becomes more complex, integrated, and differentiated, it becomes very difficult to compete with Cadence in agentic AI. Whether you are a startup or trying to develop agents for customers, the capabilities involving model routing, domain-specific models, and tight integration with agents and tools are beyond what customers can do themselves.

(00:52:37)但現在，我們在頂層進行了重大投資和差異化。我們正在定義未來所有行業的代理流。隨著我們在頂部建立代理層，並且隨著它變得更加複雜、整合和差異化，與 Cadence 在代理 AI 方面競爭變得非常困難。無論您是初創公司還是試圖為客戶開發代理，涉及模型路由、特定領域模型以及與代理和工具的緊密整合的能力都超出了客戶自己能做到的範圍。

Anirudh Devgan - Cadence, CEO

(00:53:01)This has been well recognized by our customers, who are increasingly working closely with us, not only in the middle layer but also in the top layer. This collaboration is strengthening as the agentic AI capabilities mature. Our customers realize the value of this partnership, and they are engaging more deeply with us across all layers of the technology stack. So, I feel very confident about the speed and differentiation of our Agentic offerings.

(00:53:01)這一點得到了我們客戶的充分認可，他們越來越與我們密切合作，不僅在中層，還在頂層。隨著代理 AI 能力的成熟，這種合作正在加強。我們的客戶意識到這種夥伴關係的價值，他們在技術堆棧的所有層面上與我們進行更深入的合作。所以，我對我們的 Agentic 產品的速度和差異化感到非常有信心。

Jade - Conference Operator

(00:53:50)Your next question comes from Jay Vleeschhouwer of Griffin Securities. Please go ahead.

(00:53:50)下一個問題來自於 Griffin Securities 的 Jay Vleeschhouwer。請開始。

Jay Vleeschhouwer - Griffin Securities, Analyst

(00:53:57)Thank you. Good evening. In your prepared remarks, you made important references to the components of your core EDA, particularly in digital. It is clear that over the past several years, you have gained considerable share in core EDA. However, I would like to explore the components of that share gain. Specifically, when considering the major categories within core EDA, especially digital, would it be fair to say that the single largest contributor to your dollar and share gains over recent years has been Innovus in implementation, and to a lesser extent in areas like synthesis, RTL simulation, and analysis?

(00:53:57)謝謝。晚上好。在您的準備發言中，您提到了核心 EDA 的重要組成部分，特別是在數位方面。很明顯，在過去幾年中，您在核心 EDA 中獲得了相當大的市場份額。不過，我想探討一下這一市場份額增長的組成部分。具體來說，當考慮到核心 EDA 中的主要類別，特別是數位時，是否可以公平地說，最近幾年您在收入和市場份額增長中最大的貢獻者是 Innovus，在實施方面，並在合成、RTL 模擬和分析等領域的貢獻相對較小？

Jay Vleeschhouwer - Griffin Securities, Analyst

(00:54:47)Perhaps you could elaborate on the breadth of where that incremental revenue and share have originated. In core EDA, to the extent that gains are not largely concentrated in place and route?

(00:54:47)或許您可以詳細說明這些增量收入和市場份額的來源範圍。在核心 EDA 中，增長是否並未主要集中在放置和路由上？

Anirudh Devgan - Cadence, CEO

(00:55:03)Yes, Jay, great question. Over the past five to seven years, we have gained significant ground in digital implementation, with Innovus becoming the platform of choice for advanced node design. What is particularly encouraging in the last six to twelve months is not just Innovus, but also all the tools around Innovus. I have talked for ten years now about the integration at lower nodes. As we move to three nanometer and two nanometer, it becomes difficult to separate sign-off from implementation or synthesis from implementation.

(00:55:03)是的，Jay，這是一個很好的問題。在過去五到七年中，我們在數位實施方面取得了顯著進展，Innovus 成為先進節點設計的首選平台。在過去六到十二個月中，特別令人鼓舞的是，不僅是 Innovus，還有所有與 Innovus 相關的工具。我已經談了十年關於低節點的整合。隨著我們進入三納米和兩納米，將簽署與實施或合成與實施分開變得困難。

Anirudh Devgan - Cadence, CEO

(00:55:42)With Agentic AI, there is an increased need for well-integrated vertical flows. I mentioned this last quarter as well. I am particularly pleased to see that not only are our tools in synthesis and sign-off much improved compared to a few years ago, but we are also achieving full-flow deployment with very large customers using our complete digital flow. This is a great trend for Cadence. At the same time, we maintain our undisputed leadership in analog. Okay. Additionally, the strength of Palladium, Jasper, Verisium, Agentic flows, and ChipStack is also benefiting our verification business and logic simulation.

(00:55:42)隨著 Agentic AI 的出現，對良好整合的垂直流程的需求增加。我在上個季度也提到過這一點。我特別高興地看到，我們在合成和簽署方面的工具相比幾年前有了很大的改善，並且我們也在與使用我們完整數位流程的大型客戶實現全流程部署。這對 Cadence 來說是一個很好的趨勢。同時，我們在類比領域保持著無可爭議的領導地位。好的。此外，Palladium、Jasper、Verisium、Agentic 流程和 ChipStack 的強大也在促進我們的驗證業務和邏輯模擬。

Anirudh Devgan - Cadence, CEO

(00:56:35)Overall, that is why I feel all three major parts of EDA are in a very strong position and gaining momentum. It is not just that individual tools are better and proven by large companies, but also the integration of advanced nodes with the Agentic solution. We will see how it proceeds, but overall, I feel very positive about it.

(00:56:35)總的來說，這就是為什麼我認為 EDA 的三個主要部分都處於非常強勁的地位並且正在獲得動力。不僅僅是單個工具的性能更好，並且得到了大型公司的驗證，還有先進節點與 Agentic 解決方案的整合。我們將看看這一進展，但總的來說，我對此感到非常樂觀。

Jade - Conference Operator

(00:57:02)Your next question comes from Gary Mobley of StoneX. Please go ahead.

(00:57:02)下一個問題來自於 StoneX 的 Gary Mobley。請開始。

Gary Mobley - StoneX, Analyst

(00:57:10)Hi, everyone. Good afternoon. Thank you for taking my question. Your bookings in the first quarter [期間存疑] were approximately mid-20%, around 26% year over year. If I recall correctly, 2026 was expected to be a low renewal period. So my question is: what is driving the strong backlog and bookings? Is it the use of traditional EDA tools as a result of AI agent workflows? Or is it, a reflection of the strength of the semiconductor industry? Any insights you can provide would be helpful. Thank you.

(00:57:10)大家好。下午好。謝謝您回答我的問題。您在第一季度[期間存疑]的訂單量大約是中等的 20%，年增長約 26%。如果我沒記錯，2026 年預計將是一個較低的續約期。所以我的問題是：是什麼驅動了強勁的訂單和訂單量？是因為 AI 代理工作流程導致傳統 EDA 工具的使用嗎？還是，反映了半導體行業的強勁？您能提供的任何見解都將是有幫助的。謝謝。

John Wall - Cadence, CFO

(00:57:54)Sure. Thanks, Gary. We are very pleased with the bookings performance and backlog growth. Particularly given that 2026, as you mentioned, is a comparatively lighter renewal year for us. The strength reflects healthy underlying design activity, expanded customer engagements, and strong add-on opportunities across our portfolio. We are seeing that Cadence is becoming increasingly important to customers as they accelerate their engineering roadmaps and address more complex design challenges.

(00:57:54)當然。謝謝，Gary。我們對訂單表現和訂單量增長感到非常滿意。特別是考慮到 2026 年，如您所提到的，對我們來說是一個相對較輕的續約年。這一強勁反映了健康的設計活動、擴大的客戶參與以及我們產品組閤中的強大附加機會。我們看到，隨著客戶加速他們的工程路線圖並解決更複雜的設計挑戰，Cadence 對客戶變得越來越重要。

John Wall - Cadence, CFO

(00:58:22)The backlog provides good visibility into this, but it is important to remember that backlog and bookings reflect the timing of revenue recognition, which depends on the structure and delivery of individual arrangements. So there is always a blend of timing and value in those figures. Strong bookings in a lighter renewal year are particularly encouraging because they reflect expanding customer engagement, not just the renewal calendar. So, we are very pleased with the strength of our bookings.

(00:58:22)訂單量提供了良好的可見性，但重要的是要記住，訂單量和訂單反映了收入確認的時間，這取決於個別安排的結構和交付。所以這些數字中總是有時間和價值的混合。在較輕的續約年中強勁的訂單量特別令人鼓舞，因為它們反映了擴大的客戶參與，而不僅僅是續約日曆。因此，我們對我們的訂單強勁感到非常滿意。

Anirudh Devgan - Cadence, CEO

(00:58:56)Thank you. So, Gary, needless to say, we have completed three quarters. There is still one more quarter to go, which is traditionally our strongest. Yes. So, we'll see where we end up. We are confident in you. Yes. Thank you.

(00:58:56)謝謝。所以，Gary，不用說，我們已經完成了三個季度。還有一個季度要過去，這通常是我們最強的一季。是的。所以，我們將看看最終結果如何。我們對你們充滿信心。是的。謝謝。

Jade - Conference Operator

(00:59:11)Your next question comes from Lachlan Brown of Rothschild. Please go ahead.

(00:59:11)你的下一個問題來自 Rothschild 的 Lachlan Brown。請繼續。

Lachlan Brown - Rothschild & Co Redburn, Analyst

(00:59:19)Hi, Anirudh, John. Thanks for squeezing me in. Are you seeing evidence that new AI design tools like ChipStack are lowering barriers to entry for custom silicon among non-traditional designers? And while I appreciate that logo expansion naturally takes time, are there signs that new logos could become a meaningful growth driver for TAM expansion going forward? Thanks.

(00:59:19)嗨，Anirudh，John。謝謝你讓我插入這個問題。你們是否看到新的 AI 設計工具如 ChipStack 正在降低非傳統設計師進入定製矽的門檻？雖然我理解 新客戶（logo）擴展自然需要時間，但是否有跡象表明新客戶標誌將成為未來 TAM 擴展的重要增長驅動力？謝謝。

Anirudh Devgan - Cadence, CEO

(00:59:45)Yes, absolutely. As I've said for years, as chip design becomes easier, more people will engage in it. If you step back, over the last 20 years, chip design has become about a hundred times easier; 20 years ago, companies like IBM and Intel would do CPU design with 500 people over five years. Now, a CPU design can be done in six months with about 50 people. That's roughly a hundredfold improvement. This has led to more and more designs. With AI, we expect an additional five to tenfold improvement, making it about a thousand times easier than 20 to 25 years ago.

(00:59:45)是的，絕對如此。正如我多年來所說的，隨著芯片設計變得更容易，會有更多人參與其中。如果你退一步看，在過去的 20 年中，芯片設計變得大約容易了 100 倍；20 年前，像 IBM 和 Intel 這樣的公司需要 500 人花五年時間進行 CPU 設計。現在，CPU 設計可以在六個月內由大約 50 人完成。這大約是百倍的改進。這導致了越來越多的設計。隨著 AI 的發展，我們預計會有額外的五到十倍的改進，使其比 20 到 25 年前容易約 1000 倍。

Anirudh Devgan - Cadence, CEO

(01:00:31)This will inevitably lead to increased design activity. You can see this even with hyperscalers; one or two years ago, it was unclear if they would continue designing chips. But now, companies like Google, Meta, Microsoft, Amazon, and Tesla are all increasing their design efforts. At the same time, traditional semiconductor companies are also performing well. This trend will only accelerate, especially with space companies, finance firms, physical AI, automotive companies, and robotics. The amount of design activity in automotive is remarkable, as is in robotics and related fields.

(01:00:31)這將不可避免地導致設計活動的增加。你甚至可以在超大規模公司中看到這一點；一兩年前，還不清楚他們是否會繼續設計芯片。但現在，像 Google、Meta、Microsoft、Amazon 和 Tesla 這樣的公司都在增加他們的設計努力。同時，傳統半導體公司也表現良好。這一趨勢只會加速，特別是在太空公司、金融公司、物理 AI、汽車公司和機器人領域。汽車領域的設計活動數量令人矚目，機器人及相關領域也是如此。

Anirudh Devgan - Cadence, CEO

(01:01:21)Robotics require many different chips. For example, a humanoid robot may have about 25 chips in one hand to control individual motors. I have believed this for years, and I expect to see more of it. Our job is to make these tools easier to use and provide more flexibility to our customers. For example, we did this with our agent solution in Maestro, enabling customers to do more themselves.

(01:01:21)機器人需要許多不同的芯片。例如，一個類人機器人可能在一隻手中有大約 25 個芯片來控制各個馬達。我多年來一直相信這一點，並期待看到更多。我們的工作是使這些工具更易於使用，並為我們的客戶提供更多靈活性。例如，我們在 Maestro 中使用我們的代理解決方案，讓客戶能夠自己做更多事情。

Jade - Conference Operator

(01:01:59)Your final question comes from Kelsey Chia of Citi. Please go ahead.

(01:01:59)你的最後一個問題來自 Citi 的 Kelsey Chia。請繼續。

Kelsey Chia - Citi, Analyst

(01:02:07)Hi. Hi, Anirudh and John. Congratulations on the great results. My question is about EDA. EDA growth has been very strong for the last three quarters, with 18% to 19% year-over-year growth, well above historical rates. It sounds like most of that growth is driven by hardware, and agentic AI's contribution is still quite small. Is that a fair characterization? Could you provide a sense of customer adoption levels for agentic workflows today? And is much of the revenue acceleration from agentic AI still ahead of us?

(01:02:07)嗨。嗨，Anirudh 和 John。恭喜你們取得了優異的成果。我的問題是關於 EDA 的。在過去三個季度中，EDA 的增長非常強勁，年增長率為 18% 到 19%，遠高於歷史水平。聽起來大部分增長是由硬體驅動的，而代理 AI 的貢獻仍然相當小。這是一個公平的描述嗎？你能提供今天代理工作流程的客戶採用水平嗎？而且來自代理 AI 的收入加速還在我們面前嗎？

Anirudh Devgan - Cadence, CEO

(01:02:46)Thanks for the question. Yes, we are very proud of the hardware growth. As you know, we break out recurring and non-recurring growth. As I mentioned earlier, and John also noted, our recurring growth has improved from 11% last year to 18% this year. So, that is not hardware. That's X hardware. Hardware typically appears in non-recurring growth. So, that's it. This represents a major improvement in our recurring growth from last year to this year, driven by a combination of agentic AI and the growth of underlying tools and design starts on the software side.

(01:02:46)謝謝你的問題。是的，我們對硬體增長感到非常自豪。正如你所知，我們將經常性和非經常性增長分開。正如我之前提到的，John 也指出，我們的經常性增長已從去年的 11% 提高到今年的 18%。所以，這不是硬體。那是 X 硬體。硬體通常出現在非重複性增長中。所以，就這樣。這代表我們的重複性增長從去年到今年有了重大改善，這是由於代理 AI 和基礎工具及設計啟動在軟體方面的增長的結合驅動的。

John Wall - Cadence, CFO

(01:03:39)Yeah, I can't see. I think, based on this, this is the answer. There were many questions a few minutes ago, and essentially, we believe that agentic AI is a demand accelerator for EDA. Customers have always built their own workflows around Cadence tools, and we expect them to continue developing their own agents as well. Our strategy accommodates that. What's exciting is that agents can perform much more design exploration than an individual engineer by repeatedly calling our underlying engines, creating the potential for significantly greater consumption of our core EDA tools.

(01:03:39)是的，我看不見。我認為，根據這一點，這就是答案。幾分鐘前有很多問題，基本上，我們相信代理 AI 是 EDA 的需求加速器。客戶一直圍繞 Cadence 工具建立自己的工作流程，我們預期他們也會繼續開發自己的代理。我們的策略適應了這一點。令人興奮的是，代理可以通過不斷調用我們的基礎引擎進行比單個工程師更多的設計探索，這為我們的核心 EDA 工具的消耗潛力創造了顯著的增長。

John Wall - Cadence, CFO

(01:04:11)I think you're starting to see that reflected in our EDA growth this

(01:04:11)我認為你開始在我們的 EDA 增長中看到這一點反映出來。

Jade - Conference Operator

(01:04:14)year. At this time, there are no further questions. I will now turn it back to Anirudh Devgan for closing remarks.

(01:04:14)今年。此時，沒有進一步的問題。現在我將把時間交回 Anirudh Devgan 進行結語。

Anirudh Devgan - Cadence, CEO

(01:04:27)Thank you all for joining us this afternoon. It's an exciting time for Cadence as we finish Q3 with AI-driven product leadership and strong business momentum. The Wall Street Journal ranked Cadence's board number two among all S&P 500 companies in its 2026 Best Boards list. On behalf of our employees and our board of directors, we thank our customers, partners, and investors for their continued trust and confidence in Cadence.

(01:04:27)感謝大家今天下午的參加。對於 Cadence 來說，這是一個令人興奮的時刻，因為我們以 AI 驅動的產品領導力和強勁的商業動力結束第三季度。《華爾街日報》在其 2026 年最佳董事會名單中將 Cadence 的董事會排名為所有 S&P 500 公司中的第二名。代表我們的員工和董事會，我們感謝我們的客戶、合作夥伴和投資者對 Cadence 的持續信任和信心。

Jade - Conference Operator

(01:05:09)Thank you for participating in today's Cadence Third Quarter 2026 Earnings Conference Call. This concludes today's call. You may now disconnect.

(01:05:09)感謝您參加今天的 Cadence 2026 年第三季度財報電話會議。這結束了今天的通話。您現在可以斷開連接。