![\"Arm](\"https:\/\/www.newsbeep.com\/il\/wp-content\/uploads\/2025\/09\/1757530857.jpg\")
<\/p>\n\n September 10, 2025\n <\/p>\n
\n Story\n <\/p>\n
<\/p>\n
In a recent announcement<\/a>, Arm has released its new Lumex\u00a0advanced compute subsystem (CSS) platform<\/a>, which the company says is designed for\u00a0AI experiences on flagship smartphones and next-gen PC.<\/p>\n In the blog post announcing the new platform,\u00a0Chris Bergey,\u00a0SVP and GM\u00a0of the Client Line of Business at\u00a0Arm, wrote that\u00a0Lumex unites the company’s\u00a0highest performing CPUs with Scalable Matrix Extension version 2 (SME2),\u00a0GPUs\u00a0and system IP, in order to enable\u00a0the ecosystem to bring AI devices to market faster and deliver experiences from desktop class mobile gaming to real time translation, smarter assistants, and personalized applications.<\/p>\n He also predicted that, by 2030, SME and SME2 will add over 10 billion TOPS of compute across more than 3 billion devices, delivering an exponential leap in on-device AI capability.\u00a0<\/p>\n Accoridng tothe release, Arm partners can choose exactly how they build Lumex into their SoC \u2013 they can take the platform as delivered and leverage cutting-edge physical implementations tailored to their needs, reaping time to market and time to performance benefits. Alternatively, partners can configure the platform RTL for their targeted tiers and harden the cores themselves.\u00a0<\/p>\n The platform combines:\u00a0<\/p>\n
Accelerated AI Everywhere with SME2-Enabled CPUs<\/p>\n The SME2-enabled Arm C1 CPU cluster reportedly provides dramatic AI performance gains for real-world, AI-driven tasks:\u00a0<\/p>\n
This leap in CPU AI compute enables real-time, on-device AI inference capabilities, providing users with smoother, faster experiences across interactions like audio generation, computer vision, and contextual assistants, Arm said.\u00a0\u00a0<\/p>\n For developers, AI experiences will just work on the Lumex platform.\u00a0Through the KleidiAI integration across major frameworks including\u00a0PyTorch ExecuTorch, Google LiteRT,\u00a0Alibaba MNN\u00a0and\u00a0Microsoft ONNX Runtime, apps automatically benefit from SME2 acceleration with no code changed required.<\/p>\n For developers building cross-platform apps, Lumex is designed to bring new portability, too:\u00a0<\/p>\n
Technology leaders \u2013 including Apple, Samsung, and MediaTek \u2013 are integrating AI acceleration capabilities for faster, more efficient on-device AI. Apple is powering Apple Intelligence; Samsung and MediaTek are improving responsiveness and efficiency of real-time AI applications such as translation, summarization, and personal assistants using Google Gemini.\u00a0<\/p>\n \u201cSME2-enhanced hardware enables more advanced AI models, like Gemma 3, to run directly on a wide range of devices. As SME2 continues to scale, it will enable mobile developers to seamlessly deploy the next generation of AI features across ecosystems. This will ultimately benefit end-users with low-latency experiences that are widely available on their smartphones,\u201d\u00a0said Iliyan Malchev, Distinguished Software Engineer, Android at Google.<\/p>\n Check out the full post for all the details<\/a>, but this is good news for consumer and mobile computing and th enegineers looking to get a competitive edge.\u00a0<\/p>\n
\n\tNext-generation SME2-enabled Armv9.3 CPU cluster including\u00a0C1-Ultra<\/a>\u00a0and\u00a0C1-Pro<\/a>, powering flagship devices\u00a0
\n\tNew\u00a0C1-Premium<\/a>, purpose built for the sub-flagship market, providing best in class area efficiency\u00a0
\n\tNew\u00a0Mali G1-Ultra GPU<\/a>\u00a0with next-generation ray tracing enabling advanced graphics and gaming, plus a boost to AI performance\u00a0
\n\tThe most flexible and power-aware DynamIQ Shared Unit (DSU) Arm has delivered to date: C1-DSU\u00a0
\n\tOptimized physical implementations for 3nm nodes\u00a0
\n\tDeep integration across the software stack delivering seamless AI acceleration for developers using\u00a0KleidiAI libraries<\/a>\u00a0
<\/p>\n
\n\tUp to 5x uplift in AI performance\u00a0
\n\t4.7x lower latency for speech-based workloads\u00a0
\n\t2.8x faster audio generation\u00a0\u00a0\u00a0
<\/p>\n
\n\tGoogle apps like Gmail, YouTube and Google Photos are already SME2-ready, ensuring seamless integration as Lumex-based devices hit the market\u00a0
\n\tCross platform portability means optimizations built for Android can seamlessly extend to Windows on Arm and other platforms\u00a0
\n\tPartners like Alipay are already showcasing on device LLMs running efficiently with SME2\u00a0\u00a0
<\/p>\n
![](\"https:\/\/www.newsbeep.com\/il\/wp-content\/uploads\/2025\/09\/1727191921.png\")
<\/a><\/p>\n