Featured Podcasts
Latest on X
This Valentine’s Day, let’s talk about a perfect match.💘
GPUs may get the headlines. But in the AI data center, they don’t scale alone.
As AI shifts from intense training to always-on, agent-based inference, something fundamental is changing. AI systems now run in tight
AI data centers are shifting from training jobs to always-on, agent-based inference. That change dramatically increases the need for CPUs. 🚀
Agent-based AI involves many agents running continuously, coordinating tasks, managing context, accessing data and interacting with
Arm is expanding in Austin with support from the Texas Semiconductor Innovation Fund grant!
This move will strengthen Texas’ role in global semiconductor innovation, creating 320+ jobs, new advanced lab capabilities, and building momentum for future compute and AI.👏
Arm’s share among top hyperscalers is expected to reach nearly 50% ⚡
That shift reflects a bigger reality: AI isn’t just about accelerators — CPUs power the data pipeline and system orchestration behind AI, while maximizing performance per watt at scale.
@TheFuturumGroup's
This National Apprenticeship Week, we’re proud to highlight Ayo Giwa, who completed a apprenticeship with us and has since moved into a graduate role, a brilliant example of where this pathway can lead. 👏 https://okt.to/p7em6M
Advancing on-device AI starts with extending the CPU architecture to better support machine learning workloads.
With Exynos 2600, @SamsungDSGlobal adopts Arm Scalable Matrix Extension 2 (SME2), accelerating matrix operations on CPUs and expanding the role of CPU-based AI for
Commodity infrastructure was built for a different era. 🦖
As AI training and inference scale, efficiency and system level optimization are becoming critical. Purpose-built platforms designed end-to-end are emerging as the new foundation for AI infrastructure.
