Arm debuts next-gen semi-custom Neoverse CSS N4 ‘Ranger’ platform — compute subsystem packs up to 128 cores per die on TSMC N3P
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.5 Articles
5 Articles
Arm targets agentic AI with Neoverse CSS N4
Arm has expanded its data center portfolio with the Neoverse Compute Subsystems (CSS) N4 and is building momentum around its Arm AGI CPU as the company targets infrastructure for agentic AI workloads. The two offerings give cloud providers and chip designers different routes to Arm-based AI infrastructure, from configurable custom silicon to production-ready CPUs. For […] The post Arm targets agentic AI with Neoverse CSS N4 appeared first on eeN…
Arm introduced Neoverse CSS N4, a semi-customised data center platform that can integrate up to 128 cores per die manufactured in TSMC’s N3P process. The company also confirmed that Oracle and ByteDance use their AGI chip, although it has not yet published independent performance tests.
Following the announcement of the Neoverse N3 and Neoverse V3 datacenter-class CPU architectures in 2024, Arm announced the next-generation Neoverse CSS N4 datacenter subsystem in 2026. This subsystem significantly increases the number of cores per die to 128, greatly improving core density and the number of threads. It also doubles the performance per socket and offers better energy efficiency and memory bandwidth. Furthermore, the subsystem ac…
Arm debuts next-gen semi-custom Neoverse CSS N4 ‘Ranger’ platform — compute subsystem packs up to 128 cores per die on TSMC N3P
Arm is bringing its next-gen Neoverse CSS N4 platforms to the cloud, sporting up to 128 cores per die, built on TSMC’s N3P process. Arm’s Compute Subsystem, or CSS, is a semi-custom program that allows customers to design a chip based on Arm’s IP, configuring components like core count, cache size, I/O, and connectivity to fit their specific needs. It’s the same platform we’ve seen at work everywhere from CPUs at Azure and Google Cloud to DPUs a…
[Digital Daily Reporter Kim Moon-ki] "Agentic AI workflows go beyond simply running models; complex computations such as tool invocation, database access, information retrieval, and inter-agent orchestration take place centered around the CPU. Due to issues with token costs, privacy, and latency, there is a trend toward distributing computations beyond the cloud to the edge and the physical world, and Arm, equipped with performance per watt and …
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
