Press Release

2025年7月26日至29日，2025 世界人工智慧大會將於上海世博中心與世博展覽館盛大舉行。當前生成式 AI 與大型語言模型（LLM）的發展正推動全球進入以算力驅動的全新時代，模型規模從數百億參數邁向兆級應用，驅動高效能 GPU 與先進伺服器基礎架構成為產業競爭焦點。如何在有限空間內實現高密度運算、有效散熱及能效提升，同時兼顧雲端與邊緣多場景下的彈性部署與穩定運行，已成為企業與雲端服務供應商推動 AI 落地的迫切挑戰。作為國內人工智慧領域的重要年度盛會，英業達今年不僅於自家攤位展示最新伺服器與液冷創新方案，同時于兩岸人工智慧融合發展創新展區（海峽經科區）展出多項 AI 應用解決方案，充分呈現英業達在 AI 與智慧醫療、元宇宙應用等軟硬體兼具的全面佈局，積極協助產業迎接大模型時代下的數位轉型與智慧化升級。   在英業達自家攤位（H2館-E616），將展出多款搭載最新 Intel® Xeon® 6 與 AMD 第五代 EPYC™ 處理器的高效能伺服器，涵蓋 2U 空間內多樣化配置（2U1N1P、2U1N2P、2U2N），支持 AMD、Intel等各主流平臺，協助客戶在不同算力密度需求下靈活部署。AI 運算解決方案方面，英業達將展示具備 PCIe 與 UBB2.0 GPU 模組化擴充能力的完整產品組合，其中 PCIe 介面伺服器提供 4U（支援最多 10 GPU）與 5U（支持最多 12 GPU）配置，滿足不同階段 AI 訓練與推論需求；8U及10U AI 伺服器則支援各家 UBB2.0 GPU 模組，兼具高模組化設計與散熱效率，以滿足大型模型時代對極致運算效能的需求。此外，英業達將同步展出自主研發的 46U 單相浸沒式冷卻機櫃 Lambert，專為高功耗、高密度部署環境設計，提供高效率液冷散熱方案，以及邊緣浸沒冷卻伺服器一體機 Nobel，專為邊緣計算應用打造，將冷卻泵直接整合於伺服器本體，採用空氣到液體冷卻方式，無需冷卻塔或冰水機組，實現真正的一體式冷卻方案，不僅能大幅提升能效比，亦有效簡化安裝與部署流程，協助使用者在空間有限或環境嚴苛的邊緣場景中有效部署 AI 運算。   于兩岸人工智慧融合發展創新展區（H2館-E116），英業達將展示多項 AI 與智慧醫療、元宇宙應用的創新成果，呈現 AI 技術在醫療、生活及各個產業中的實際應用價值。展出內容包括運用深度學習技術進行影像分析、協助醫師快速判讀傷口病灶的「英愛膚 I-SWAT 智慧傷口軟體」，適用於急診室檢傷、慢性病患遠端照護及偏遠地區醫療協助，提升醫療效率與可及性；「myHeartScore」作為一款創新健康APP，其應用程式整合 Apple Watch 所搜集的心電圖(ECG)與心率變異性(HRV)資料，進行即時且個人化的風險分析，透過雙 AI 模型能於 30 秒內預測心血管風險，並進行每日追蹤與長期健康監測，協助使用者及早掌握心血管健康變化並預防潛在風險；此外，英業達于展會更展出最新的腸胃道AI解決方案，針對腸胃道癌症的預防和檢測，提高了內視鏡檢查的準確性和效率，並且整合高效能運算應用程式，提升即時運行效能，確保系統能夠在精准且嚴苛的醫療環境中可靠地運行先進的AI模型，從而為臨床醫生及其患者提供更安全、更強大的支援。   另外英業達旗下元宇宙戰略品牌、虛實融合行業先鋒「虛實元邦 VRSTATE」也將在展區內重點呈現基於「以實推虛、以虛帶實、虛實共生、生生不息」核心理念開發的「領域智慧體（Domain Agents）」解決方案，藉此展示利用AI x Metaverse賦能千行百業的前沿解決方案，帶來創新數字經濟圈的發展，並且在本次大會展館設置了專門的「領域智慧體體驗專區」，觀眾將能通過邊緣智算工作站及觸控屏等設備來與Domain Agents(元邦健管師、元邦導覽員、元邦時尚家等)進行智慧互動式體驗，專區還將預告二次元動漫化身，並設有裸眼3D設備展示區，全方位展示 VRSTATE 在 AI x Metaverse 領域的領先技術。   誠摯邀請您蒞臨上海世博展覽館 📍 H2館-E616 英業達展位 📍 H2館-E116 兩岸人工智慧融合發展創新展區（海峽經科） 與英業達專家深入交流，共同探索 AI 與資料中心前沿技術與產業最新趨勢，攜手迎接智慧時代的無限可能。   與英業達專家深入交流，共同探索 AI 與資料中心前沿技術與產業最新趨勢，攜手迎接智慧時代的無限可能。

2025年6月11日至13日，第十一屆上海國際數據中心產業展覽會將於上海新國際博覽中心盛大舉行。作為國內數據中心產業的重要年度盛會，英業達深感榮幸受邀參展，並將於現場展示多項創新技術與最新解決方案。   隨著AI生態系的迅速發展，整個產業正面臨前所未有的挑戰與轉型。本次展會聚焦多項關鍵議題，包括高算力需求驅動下的大模型訓練與部署、高功率設備所帶來的液冷散熱挑戰、自動化運維技術的快速導入，以及大型園區數據中心對新型能源供應模式的探索，這些皆是推動產業升級的重要方向。   英業達深耕伺服器領域近三十年，始終專注於高效能運算、AI伺服器、邊緣計算與先進冷卻技術的研發，致力於為各種應用場景提供量身打造的整體解決方案。在本屆展會中，英業達將展出多款搭載最新 Intel® Xeon® 6 處理器與 AMD 第五代 EPYC™ 處理器的高效能伺服器，產品涵蓋多種2U空間內的靈活配置，如2U1N1P、2U1N2P與2U2N架構，可支援不同規模與算力密度的部署需求，並全面囊括AMD、Intel與NVIDIA等主流CPU平台。   AI伺服器方面，英業達將展示多款具備PCIe GPU與UBB2.0 GPU模組解決方案的高擴展性服務器。其中，搭載PCIe介面GPU的AI伺服器提供4U與5U兩種產品尺寸，分別支援最多10張與12張GPU，適用於不同的AI訓練與推論場景。針對更高算力需求，英業達亦推出支援UBB2.0 GPU模組的8U AI伺服器，具備高度模組化與散熱優化設計，並可相容各家UBB2.0 GPU模組解決方案，滿足資料中心在AI時代的運算效能與彈性擴充需求。   此次展出亮點之一，是由英業達自主研發的46U單相浸沒式冷卻機櫃 Lambert，專為高密度、高功耗應用打造，具備高度整合的冷卻架構。另一項重要展示為首次亮相的邊緣浸沒冷卻伺服器一體機 Nobel，專為邊緣場景應用設計，採用高容量小型單體架構，將冷卻泵直接整合於伺服器內部，實現真正的一體式冷卻方案。Nobel 採用空氣到液體的散熱方式，無需依賴冷卻塔或冰水機組，不僅能大幅提升能效比，亦有效簡化安裝與部署流程，特別適合空間受限或環境條件嚴苛的邊緣應用環境。   誠摯邀請您蒞臨上海新國際博覽中心 N4館667號英業達攤位，與我們的專家深入交流，共同探索數據中心產業的未來趨勢與前沿技術。

Redefining intelligent computing, from AI servers and Edge AI to Metaverse applications and digital creativity.  Taipei, Taiwan — May 19, 2025 — Inventec Corporation, a global leader in technology innovation, is set to participate in COMPUTEX 2025, taking place from May 20 to 23 at the Nangang Exhibition Hall in Taipei. Committed to constant research and innovation, Inventec continues to provide forward-looking solutions designed to meet the needs of customers in an ever-evolving digital landscape.  At this year’s event, Inventec will present a comprehensive range of high-performance AI servers, intelligent Edge AI solutions, diversified AI applications, next-generation automotive electronics, innovative Metaverse application platforms, and its intelligent virtual artist, Mila Mii, delivering a unique blend of technology and creativity.  Pushing AI Computing Boundaries with Next-Generation Servers One of the highlights at the Inventec booth will be the Artemis II rack solution, developed in collaboration with NVIDIA. Built on NVIDIA GB300 NVL72 system and adopting the NVIDIA MGX platform, Artemis II is designed to support the intensive compute demands of large-scale generative AI training and inference workloads. This collaboration reflects Inventec’s commitment to delivering future-ready infrastructure for AI-driven data centers.  Inventec will also showcase a lineup of versatile AI servers tailored to meet modern computational demands. These include servers equipped with NVIDIA Blackwell GPU architecture and Intel Xeon Scalable Processors, such as the P4000 with NVIDIA RTX PRO 6000 Blackwell Server Edition and Intel Xeon 6700/6500 Series; the P9000G7, integrating NVIDIA Blackwell Ultra  GPUs and both Intel Xeon 6 processors with P-core and E-core; and the P9000G6 and P9000LG6, which feature different cooling solutions but both support NVIDIA Blackwell GPUs along with 4th and 5th generation Intel Xeon Scalable Processors. All of these AI servers also serve as key infrastructure platforms for the NVIDIA Enterprise AI Factory concept. The NVIDIA Enterprise AI Factory validated design provides guidance for developing, deploying, and managing agentic AI, physical AI, and HPC workloads on the NVIDIA Blackwell platform on-premises environments.  Beyond NVIDIA-based configurations, Inventec will also introduce the P8500G6 AI server, designed to support AMD Instinct MI300 Series GPUs, as well as the latest AMD GPU solutions, offering organizations enhanced flexibility and scalable performance for their AI workloads.  Empowering the Edge and Intelligent Industries Recognizing the rising importance of real-time AI computing at the edge, Inventec will showcase its Edge AI solutions at COMPUTEX. These systems are designed to meet the demands of a wide range of applications, including smart factories and intelligent transportation, and are built around NVIDIA Blackwell GPUs and NVIDIA Jetson to deliver efficient performance for diverse industry scenarios. Additionally, Inventec will showcase the NX-R2028, a compact, low-power private 5G network device designed for smart factories, offering easy maintenance and seamless deployment for industrial environments.  Furthermore, Inventec will highlight its smart retail solution developed in collaboration with SYSCOM. The Agentic AI system, powered by edge computing devices, recommends products based on customer profiles, and then it’s streamlined with a robot prompting shoppers with tailored offers and product location info. This interactive process could boost retail sales and become a part of future commerce.  Bringing AI Innovation to Everyday Life and the Metaverse Beyond hardware, Inventec will also highlight a portfolio of AI-powered applications across smart factories, intelligent transportation, smart cities, healthcare, and the Metaverse. One of the featured solutions, MyHeartScore, empowers users to monitor cardiovascular health, detect potential risks, and stay proactive about their well-being using AI-driven insights.  Following the 2025 commercial launch of its AI wound classification software, I-SWAT, Inventec is now expanding into gastrointestinal cancer prevention and detection — a key healthcare need in Taiwan. The Gastrointestinal AI enhances the accuracy and efficiency of endoscopic exams, with Holoscan providing real-time sensor processing for ultra-low latency endoscopic imaging. By integrating NVIDIA IGX and Holoscan, Inventec ensures the system runs advanced AI models reliably in precise and demanding medical environments, delivering safer and more robust support for clinicians and their patients.  To complement its technological showcase, Inventec’s intelligent virtual artist, Mila Mii, developed using NVIDIA Audio2Face-3D technology, will perform live throughout the four-day exhibition. Mila’s lifelike expressions and emotionally rich vocals will offer visitors an inspiring demonstration of how AI can blur the boundaries between art and technology.  Inventec warmly invites all industry partners, clients, and visitors to experience firsthand how its solutions are shaping the future of AI-driven innovation at the Inventec booth I0617a at COMPUTEX 2025.

Inventec Corporation is pleased to participate as a Gold Sponsor at NVIDIA GTC 2025, the global AI industry event held from March 18-21 at the San Jose Convention Center.  The company will showcase its comprehensive capabilities, from edge to rack-level AI solutions, designed to support high-performance computing and AI applications.    At GTC, Inventec is excited to debut an array of AI server innovations, including the P4000G7, an Inventec-built 4U NVIDIA MGX server that delivers exceptional performance while adhering to NVIDIA MGX specifications. It is equipped with NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, a powerful data center GPU for AI and visual computing. Additionally, the P9000IG6 and P9000IG6L offer versatile AI server configurations ranging from 8U to 10U, designed to accommodate various cooling requirements and equipped with the latest NVIDIA Blackwell GPUs. Inventec will also unveil a series of edge AI devices powered by the NVIDIA Jetson platform, designed to address diverse industry applications. The key highlight will be Artemis II, an AI rack solution. Built with the NVIDIA Blackwell Ultra GB300 NVL72 rack-scale solution that leads the new era of AI and the next wave of accelerated computing, providing a scalable AI computing platform capable of meeting the ever-evolving demands of modern workloads.    Beyond product innovations, Inventec will also share its expertise through two key sessions at GTC. The first, titled "Manufacturing for AI and AI for Manufacturing," will be available as an on-demand session on the GTC portal, offering insights into how Inventec leverages NVIDIA technology to build high-performance AI servers and develop AI-driven smart manufacturing solutions. The second session, "AI Agent-in-a-Box: Design and Deploy Robust Edge AI Applications," will take place on Wednesday, March 19, from 4:00 PM to 4:15 PM PDT in the exhibit hall. This presentation will share insights into developing AI applications on high-performance servers and deploying solutions using edge AI boxes powered by the NVIDIA Jetson platform.    Inventec warmly welcomes all attendees to visit booth #746 to experience its latest AI server and device solutions firsthand and engage with its team of experts. This event presents an excellent opportunity to gain deeper insights into the future of AI infrastructure and edge computing