Arm recently announced the arrival of its Lumex CSS platform, which it says is designed to support next-generation AI features on consumer devices. That in mind, the new platform combines powerful CPUs, GPUs, and system IP with an optimized software stack.
READ: Qualcomm Teams up with BMW for a AI-Enabled Automated Driving System
One of Lumex’s core components is the Scalable Matrix Extension version 2 (SME2), which Arm says enhances AI performance by enabling new SME2-enabled CPUs to achieve up to a 5x increase in AI performance. On paper, this essentially translates to faster real-time tasks like voice translation and audio generation, for example. Chris Bergey, SVP and GM of Arm’s Client Line of Business states:
AI is no longer a feature, it’s the foundation of next-generation mobile and consumer technology. Users now expect real-time assistance, seamless communication, or personalized content that is instant, private, and available on device, without compromise. Meeting these expectations requires more than incremental upgrades, it demands a step change that brings performance, privacy and efficiency together in a scalable way.
By moving AI processing from the cloud to the device, Arm says that Lumex is able to provide users a faster, more private, and more reliable experience. It also supports graphics enhancement and gaming with the new Mali G1-Ultra GPU, with as much as a 2x uplift in ray tracing performance and up to a 20% increase in AI inference performance.
For developers, the KleidiAI software library is integrated into major mobile operating systems and AI frameworks, allowing them to easily access SME2 acceleration. Arm says that this technology will add over 10 billion TOPS of compute across more than 3 billion devices over the next half-decade.