本網站使用瀏覽器紀錄 (Cookies) 來提供您最好的使用體驗，我們使用的 Cookie 也包括了第三方 Cookie。相關資訊請訪問我們的隱私權與 Cookie 政策。如果您選擇繼續瀏覽或關閉這個提示，便表示您已接受我們的網站使用條款。 關閉
TAIPEI, Jan. 16, 2025 /PRNewswire/ -- Aetina, a leading provider of edge AI solutions, announces its comprehensive support for Super Mode, a feature introduced in the latest NVIDIA® Jetson Orin™ NX and Jetson Orin™ Nano modules. This enhancement, set to be enabled by the NVIDIA JetPack™ software development kit's 6.2 release this month, delivers up to 2x boost in generative AI inference performance, reinforcing powerful applications in computer vision, robotics, and local AI deployment. Super Mode transforms the operational capabilities of the Jetson Orin platforms by enhancing their processing performance and efficiency, making them ideal for demanding AI applications at the edge. With Super Mode, Jetson Orin NX and Orin Nano modules deliver up to a 70% increase in AI TOPS, allowing a wide spectrum of AI models to run, including LLMs, VLMs, and Vision Transformers (ViTs). Aetina's DeviceEdge series, including models AIE-CN/CO-1-S1, AIE-PN/PO-1-S1 and AIE-KN/KO-1-S1, are fully optimized to leverage these substantial enhancements, providing a formidable platform for enterprise and industrial AI solutions. Key Advancements with Super Mode: Boost AI Capabilities: Super Mode enables more complex generative AI models to run faster and more efficiently, ideal for versatile applications potential, including LLM chatbot, visual AI agent, AI-based robotics and proactive edge computing applications. Comprehensive Ecosystem Support: The rich NVIDIA software ecosystem combines essential development tools like NVIDIA Isaac, NVIDIA Metropolis, and NVIDIA TAO Toolkit, enabling developers to rapidly build and deploy sophisticated AI solutions with confidence. "As an Elite member of the NVIDIA Partner Network, Aetina is committed to bringing cutting-edge AI capabilities to the edge computing market," said Troy Lin, Senior Manager of Product Development at Aetina. "The integration of Super Mode for Jetson Orin NX and Orin Nano modules into our DeviceEdge series represents a significant leap forward in edge AI computing. It is more than just a performance upgrade - it's a strategic investment in future AI capabilities. Super Mode helps ensure our clients' AI infrastructure stays ahead of emerging technologies and future workload demands." With the NVIDIA JetPack 6.2 release, Aetina is poised to propel its clients into a new era of AI with enhanced capabilities that promise to revolutionize industrial and enterprise applications. Alongside the software updates, Aetina will launch new system models with Super Mode for Jetson Orin Nano modules in Q1 and Super Mode for Jetson Orin NX modules in Q2 2025 to accommodate the enhanced performance needs, ensuring system durability and reliability across a variety of operational conditions.
LAS VEGAS, Jan. 10, 2025 /PRNewswire/ -- Youdao, a leading intelligent learning company in China, is showcasing its AI-equipped smart device, the Youdao Dictionary Pen X7 series, at the International Consumer Electronics Show (CES 2025).This event showcases the strength of 'Chinese AI' in the field of educational technology to a global audience. The Youdao booth attracted significant attention at the exhibition. The showcased Youdao Dictionary Pen X7 series features proprietary applications, including Hi Echo and Mr. P AI Tutor, which are based on the 'Zi Yue' educational large language model. It also provides offline translation capabilities through an edge-side large language model, drawing many users to experience the products. During the exhibition, Youdao was invited to participate in the Microsoft Greater China offline event at CES 2025. Renlei Liu, Senior Vice President of Youdao, attended a high-level roundtable discussion. Alongside representatives from tech companies such as Microsoft, TCL, Lenovo, and TimeKettle, they explored AI innovation and the globalization of consumer electronics in China. Renlei Liu stated that large language models possess exceptional natural language processing and interaction capabilities. Leveraging its extensive experience in the language and learning sectors, Youdao launched China's first educational large language model, 'Zi Yue', in 2023. Based on this model, Youdao upgraded all of its software and hardware products and introduced a series of proprietary educational applications, including the digital human language coach, AI Document Q&A, and transcription services for international students. Renlei Liu said, 'We believe that before the emergence of truly super applications based on large language models, various companies are exploring vertical fields and specific scenarios, much like the early days of China's mobile internet, when killer applications had not yet emerged and internet companies were innovating within vertical scenarios. There are many similarities between the era of large models and the mobile internet era, and we are actively experimenting as well'. Youdao has also applied the reasoning and knowledge Q&A capabilities of the large language model to academic learning scenarios, launching the AI comprehensive inquiry learning assistant, Mr. P AI Tutor. It integrates the large language model and RAG (Retrieval-Augmented Generation) technology, enhancing the AI's understanding of subject knowledge and the quality of its outputs. It offers guided explanations, multi-turn interactive heuristic teaching, and comprehensive Q&A support across all grades and subjects, addressing many issues in home tutoring scenarios. Renlei Liu stated that Youdao is actively promoting the application of the "Zi Yue" educational large language model in various scenarios and has conducted a comprehensive AI upgrade of its business products to better meet users' learning needs. As a result, Youdao's AI subscription service has achieved rapid growth for seven consecutive quarters. With the continuous development of large language model applications, improving the efficiency of AI research and development and application has become a key factor in industry competition. In response, Youdao has proposed a solution to harness the capabilities of smaller model applications. Renlei Liu pointed out that general large language models often aim for "large and comprehensive," but the expansion of model size also leads to increased training and operational costs. For specific scenarios and tasks that require quick response times, using smaller models on the edge can facilitate more effective targeted training, enhance precision, and provide a better user experience. The AI dictionary pen, equipped with a dedicated educational model can provide high-quality content and functional services, ensuring that it remains focused on learning purposes and is not used for entertainment or other applications. At the same time, by utilizing a smaller model on the edge for localized operation, it supports offline use, thereby avoiding disruptions from the network environment. Renlei Liu stated that smaller models focus on high-quality vertical applications, which can continuously enhance user experience and build professional technical advantages through flexible and efficient training and iteration. Exploring the application of edge models in more specific scenarios is an important direction for Youdao's development in AI technology. In fact, Youdao has already made significant progress in the exploration of edge models. Based on its self-developed "Zi Yue" educational large language model, Youdao has developed the industry's first edge large language model and applied it to the newly launched X7 series of dictionary pens. This dictionary pen features offline large language model translation capabilities, achieving translation results comparable to those of online NMT (Neural Machine Translation), greatly enhancing the accuracy and fluency of language learning. As a star product in Youdao's smart devices lineup, the Youdao Dictionary Pen X7 series has truly become an AI large language model dictionary pen. It is equipped with two proprietary applications of the first educational large language model, "Zi Yue" in China: the AI All-Subject Tutor, Mr. P AI Tutor, and the digital human language coach, Hi Echo, while also providing AI intelligent learning features such as AI essay refinement. The Youdao Dictionary Pen X7 received an enthusiastic response from users upon its launch. It quickly became the best-selling product in the thousand-yuan price range, surpassing 30,000 units sold. During the 2024 Nov 11th Shopping Festival in China, the Youdao Dictionary Pen X7 Pro topped several industry rankings on major e-commerce platforms, marking the fifth consecutive year that Youdao dictionary pens have held the top position for sales in the same category across both JD.com and Tmall. This time, Youdao showcased the Dictionary Pen X7 series at CES 2025, receiving widespread recognition on an international stage. Focusing on AI + education, Youdao will continue to increase its investment in the research and development of edge model applications, promoting the implementation of large language models in more educational scenarios to provide learners with a better learning experience through efficient educational products. About Youdao, Inc. Youdao, Inc. (NYSE: DAO) is a leading intelligent learning company in China focused on using technology to provide educational content, applications, and solutions for learners of all ages. Founded in 2006 as part of NetEase, Inc. (NASDAQ: NTES; HKEX: 9999), Youdao offers a wide range of online courses, interactive learning apps, and smart learning devices. For more information, please visit: http://ir.youdao.com & https://smart.youdao.com/en/. CONTACT: youdaopr@rd.netease.com
TAIPEI, Jan. 10, 2025 /PRNewswire/ -- Aetina Corporation, a leading provider of edge AI solutions and a subsidiary of Innodisk Group, today announced a collaboration with Qualcomm Technologies, Inc., who unveiled a revolutionary Qualcomm® AI On-Prem Appliance Solution and Qualcomm® AI Inference Suite for On-Prem. This collaboration combines Qualcomm Technologies' cutting-edge inference accelerators and advanced software with Aetina's edge computing hardware to deliver unprecedented computing power and ready-to-use AI applications for enterprises and industrial organizations. Aetina Advances Edge AI with Qualcomm® Cloud AI Family of Accelerator Cards The flagship offering, the Aetina MegaEdge AIP-FR68, sets a new industry benchmark by integrating Qualcomm® Cloud AI family of accelerator cards. Each Cloud AI 100 Ultra card delivers an impressive 870 TOPS of AI computing power at 8-bit integer (INT8) while maintaining remarkable energy efficiency at just 150W power consumption. The system supports dual Cloud AI 100 Ultra cards in a single desktop workstation. This groundbreaking combination of power and efficiency in a compact form factor revolutionizes on-premises AI processing, making enterprise-grade computing more accessible than ever. Richard Hung, Vice President at Aetina, emphasized the transformative potential: "Our collaboration with Qualcomm Technologies marks a significant breakthrough in edge AI deployment. By achieving the complex landscape in a desktop form factor, we're democratizing access to enterprise-grade AI capabilities. This solution enables businesses to run sophisticated AI models locally, ensuring data privacy while maintaining superior interoperability and cloud-level performance." The collaboration introduces innovative applications in generative AI, particularly through Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG). This advancement enables organizations to rapidly develop, test, and customize generative AI models on standard PC workstations, with seamless scalability to private or cloud data centers. This capability is particularly valuable for industries requiring secure, high-performance AI processing, such as healthcare, finance, retail and manufacturing. Additionally, the collaboration strengthens Aetina's role as a key AI innovator within the Innodisk Group, highlighting its expertise in global partnerships. "We are excited to introduce novel and advanced on-prem AI inference appliance solutions with Aetina," said Rashid Attar, Senior Vice President of Engineering, Qualcomm Technologies, Inc. "These solutions address enterprise on-prem AI needs across a diverse set of sectors and combined with the AI Inference Suite enable a broad range of AI agents, applications, and use cases." The Qualcomm AI On-Prem Appliance Solution combines state-of-the-art processors with comprehensive Qualcomm AI Inference Suite for On-Prem to offer a powerful, localized AI computing solution that operates efficiently within enterprise environments. The new bundle solution will be available through global distribution partners starting in Q1 2025, marking a significant milestone in making enterprise-grade AI accessible to organizations of all sizes. For more information on how Aetina is revolutionizing edge computing, please visit Aetina's website. About Aetina Corporation Aetina, a provider of state-of-the-art AI solutions, offers a wide range of AI computing systems, platforms, hardware, and software tools ideal for the creation of different types of vertical AI. Aetina's integrated solutions empower its Arm, x86 computers, GPUs, ASIC hardware, cloud management software, and development tools with artificial intelligence, delivering comprehensive, tailor-made hardware and software suites to enable smooth, quick system intellectualization progress at the edge. With its ecosystem network, Aetina makes the creation and deployment of any specific AI-powered applications highly achievable by leveraging its global partners' AI technologies.Read more at: www.aetina.com Qualcomm is a trademark or registered trademark of Qualcomm Incorporated. Qualcomm branded products are products of Qualcomm Technologies, Inc. and/or its subsidiaries.
Innovative VOCCLEAR® Technology, Tinnitus Relief, and Real-Time Translation Take Center Stage at ELEHEAR's Demos at the Global Conference LAS VEGAS, Jan. 4, 2025 /PRNewswire/ -- ELEHEAR, a global leader in AI-powered smart hearing wearables, is set to make waves at CES 2025 with its industry-leading hearing enhancement technology. Following the successful launch of the ELEHEAR Beyond smart hearing wearable this past fall—which introduced advanced AI features to the market at an accessible price—the company is excited to unveil its latest innovation, ELEHEAR Beyond Pro, at the world's largest tech event. Building on the success of the ELEHEAR Beyond—heralded as "cutting-edge hearing technology within reach"—ELEHEAR is redefining accessibility, usability, and functionality in the AI-powered smart hearing wearable space and aims to redefine the category as functionality expands to include AI integrations making ELEHEAR devices a quintessential every-day-carry necessity. With its proprietary VOCCLEAR® technology, ELEHEAR's smart hearing devices deliver exceptional sound clarity, with a customizable app and setup that is super easy to navigate. These aren't your average hearing devices—they're the ultimate fusion of health tech and modern innovation. ELEHEAR will also show a sneak peak of the ELEHEAR Beyond Pro, which is to be released later this year, and will feature a new Champagne color along with several upgrades for enhanced sound clarity and an improved user experience. CES attendees will be among the first to experience the forthcoming new device. "Our mission with ELEHEAR Beyond is simple," says Eric Miao, CEO of ELEHEAR, "to deliver cutting-edge hearing technology at a price everyone can afford. With advanced AI-driven features, such as VOCCLEAR® and noise cancellation, we're giving users the premium hearing experience they deserve at a fraction of the traditional cost." He added "CES 2025 gives us the platform to show the world how ELEHEAR is leading innovation in the smart hearing wearables market and interact directly with our partners and customers." What to Expect at CES 2025 ELEHEAR's Beyond model features real-time AI translation that supports live translations across multiple languages, pushing past traditional hearing aid capabilities and setting a new standard for AI integration in the field. With 20 customizable masking sounds, ELEHEAR smart hearing wearables provide personalized tinnitus relief, enhancing users' well-being and enabling effortless communication. Seamlessly stream music and calls from iOS or Android devices with Bluetooth 5.3 connectivity, transforming Elehear hearing aids into premium wireless headphones. Through the ELEHEAR app, users can personalize their hearing experience with personalized compensation adjustments, volume controls, and over-the-air firmware updates for continuous performance improvements. Award-Winning RecognitionELEHEAR's innovative approach has been celebrated with the Expert Choice Award 2024 from Hear Advisor, as well as recognition from Soundly, WIRED, Forbes Health, and NCOA. Its products have been praised for offering premium technology at a fraction of the traditional cost, making hearing aids more accessible to millions with its user centric design and user interfaces and have been awarded iF and Red Dot design awards. The Future of Hearing at CESVisitors to CES 2025 can experience ELEHEAR's latest technology firsthand at CES Unveiled on Sunday, January 5th, at Mandalay Bay (5–8:30 PM), PEPCOM Digital Experience on Monday, January 6th, at Caesars Palace (7–10:30 PM), or visit their suite on the CES Show Floor at the Venetian Tower, Hospitality - Venetian Palazzo Hospitality Suites. From live demonstrations of real-time AI translation to interactive app tutorials, attendees will see how ELEHEAR is transforming hearing aids into multifunctional, high-tech devices for everyday use. About ELEHEAR ELEHEAR's technology comes from a diverse team of innovators hailing from the U.S., Canada, Europe, and China. ELEHEAR has established strategic partnerships with global tech leaders and research institutes. Its proprietary technologies have amassed extensive real-world usage experience across various consumer electronics products and have gained recognition in both academic and industrial fields. ELEHEAR Beyond builds on this foundation, bringing expertise from the consumer tech space directly into the hearing aid market. With this launch, ELEHEAR is set to empower millions of hearing users to enjoy better hearing, improved communication, and a higher quality of life. Website: www.elehear.com
TAIPEI, Jan. 3, 2025 /PRNewswire/ -- DFI, a global leader in embedded motherboards and industrial computers, will join DEEPX at CES 2025 to unveil new industrial edge AI platforms integrated with the DX-M1 AI accelerator, targeting smart city applications. This collaboration reflects a shared commitment to advancing on-device and application-specific AI, paving the way for a more flexible, efficient, and sustainable edge AI adoption. DFI and DEEPX Showcase Edge AI Solutions at CES 2025 According to Research and Markets, the global edge AI market is projected to grow at a 20.54% CAGR, increasing from US$23.571 billion in 2024 to US$59.978 billion by 2029. To meet the growing demand for scalable and energy-efficient AI solutions, DFI integrates the DEEPX DX-M1 AI accelerator into its industrial hardware platforms, delivering cutting-edge performance tailored to modern industry needs. The DEEPX DX-M1 excels with the ability to process over 16 video channels on a single chip, achieve real-time performance exceeding 30 fps with GPU-level precision, and deliver up to 25 TOPs of AI power while minimizing energy consumption. This combination of power and efficiency enables simultaneous execution of advanced AI algorithms, such as object recognition and image classification, making it a versatile solution for applications in industrial robotics, machine vision, AI-driven IPC and HPC, smart factories, and beyond. To demonstrate this vision, DFI will showcase two fan-less embedded systems—the EC710-ASL and EC600-RPS—featuring the DEEPX DX-M1 M.2 AI accelerator. These platforms are designed to meet critical real-time demands in smart cities, enabling applications like smart transportation, surveillance, and accident prevention. The system are powered by Intel® Atom® and Intel® 14th/13th/12th Gen processors, and support Intel® Time Coordinated Computing (Intel® TCC) technology for precise real-time operations while maintaining low power consumption and minimal latency. DFI continues to push the boundaries of industrial-grade embedded solutions with its latest product innovations, including the ASL253 single-board computer powered by Intel® Atom® processors, the EB100-MTU NUC system featuring Intel® Core™ Ultra processors, proprietary Out-of-Band (OOB) modules, and the versatile DFI Workload Platform. These advancements not only boost hardware performance but also enhance remote management through OOB capabilities and optimize AI efficiency to meet the growing demands of edge AIoT applications. Visit DFI at the DEEPX booth (#9045) during CES 2025 (January 7–11) in the North Hall, Las Vegas Convention Center, to explore our edge AI platform designed for smart city applications.
TAIPEI, Nov. 20, 2024 /PRNewswire/ -- Aetina, a subsidiary of the Innodisk Group and an expert in edge AI solutions, is pleased to announce its debut at Supercomputing (SC24) in Atlanta, Georgia, showcasing the innovative SuperEdge NVIDIA MGX™ short-depth edge AI server, AEX-2UA1. By integrating an enterprise-class on-premises large language model (LLM) with the advanced retrieval-augmented generation (RAG) technique, Aetina NVIDIA MGX™ short-depth server demonstrates exceptional enterprise edge AI performance, setting a new benchmark in Edge AI innovation. The server is powered by the latest Intel® Xeon® 6 processor and dual high-end double-width NVIDIA GPUs, delivering ultimate AI computing power in a compact 2U form factor, accelerating Gen AI at the edge. Aetina SuperEdge NVIDIA MGX™ Short-Depth Edge AI Server The SuperEdge NVIDIA MGX™ server expands Aetina's product portfolio from specialized edge devices to comprehensive AI server solutions, propelling a key milestone in Innodisk Group's AI roadmap, from sensors and storage to AI software, computing platforms, and now AI edge servers. NVIDIA MGX™ Edge AI Server Showcases Enterprise-Class Gen AI To accelerate Generative AI for enterprises, Aetina provides the compact and high-performance AEX-2UA1 NVIDIA MGX™ short-depth server, a single socket server integrating the Intel® Xeon® 6 processor. By supporting the dual double-width NVIDIA GPUs with ultra-fast NVIDIA NVLinkTM GPU interconnect, the AEX-2UA1 unleashes ultimate edge computing performance in a compact 2U form factor. Additionally, it supports advanced networking with NVIDIA BlueField®-3 DPUs and NVIDIA ConnectX®-7 NICs for high-speed communication. The AEX-2UA1 empowers enterprises to foster the private LLM, ensuring secure access to sensitive data while meeting stringent regulatory compliance. Built on a modular architecture, the AEX-2UA1 also offers remarkable flexibility and compatibility with current and future NVIDIA hardware, delivering greater scalability for enterprises in the future. Aetina x Innodisk: Architect Intelligence in the Edge The announcement of NVIDIA MGX™ server reaffirms Aetina's position as the driving force in AI innovation within the Innodisk Group, showcasing its strength and momentum in pushing technological boundaries. Alongside Innodisk, we facilitate the widespread adoption of edge AI across diverse markets, enhancing the transformation of cutting-edge innovations into practical endpoint solutions. Aetina showcases comprehensive edge AI solutions from AI inference systems to AI training servers at SC24. Visit Aetina (Booth No. 420) to discover how our innovative edge AI server can bring business operations to the next level. For more information about SuperEdge short-depth server, please visit https://www.aetina.com/ .
A12 藝術空間
Edge AI
請先登入後才能發佈新聞。
還不是會員嗎？立即 加入台灣產經新聞網會員 ，使用免費新聞發佈服務。 （服務項目） （投稿規範）
