Arm Holdings
-
Moving AI Workloads from Cloud to On-Premise: A Strategy for Reducing Power Consumption
Arm CEO Rene Haas advocates for distributing AI workloads from cloud-based infrastructure to local devices to reduce energy consumption and improve sustainability. He highlights a shift towards hybrid computing, with AI training in the cloud and inference occurring on devices like smartphones and AR glasses. Arm’s expanded partnership with Meta aims to optimize AI efficiency across the entire compute stack, exemplified by localized speech recognition in Meta’s Ray-Ban Wayfarer glasses. This localized processing enhances responsiveness and reduces reliance on cloud servers.
-
Arm GPUs to Add Dedicated Neural Accelerators in 2026, Devices Arriving Next Year
Arm Holdings plans to launch its dedicated Neural Processing Unit (NPU) for GPUs in 2026, with devices expected by late next year. Simultaneously, Arm unveils the open-source Neural Graphics Development Kit to streamline integration. A key application is Neural Super Sampling (NSS), which upscales resolution with low latency and near-native quality, reducing GPU workload by up to 50%. This innovation promises improved power consumption, higher frame rates, and enhanced visuals for gaming and other applications.
