Arm Unveils Its First Data Center Chip
Arm unveiled its first in-house data center processor, shifting away from licensing chip designs for other companies to use their own production silicon.

Arm unveiled its first in-house data center processor, shifting away from licensing chip designs for other companies to use their own production silicon. The chip pits Arm against its customers, who also make AI chips for data centers. The company’s first silicon is the AGI CPU, a data-center processor positioned for use across agentic AI infrastructure.
CEO Rene Haas said AI is reshaping how computing is designed and delivered, noting agentic computing is speeding the shift. “Today marks the next phase of the Arm compute platform and a defining moment for our company. With the expansion into delivering production silicon with our Arm AGI CPU, we are giving partners more choices, all built on Arm’s foundation of high-performance, power-efficient computing, to support agentic AI infrastructure at a global scale.”
The chip packs up to 136 Neoverse V3 cores per CPU, delivers 6GB/s of memory bandwidth per core at sub-100-nanosecond latency, and has a thermal design power of 300 watts. Arm stated the AGI CPU delivers more than double the performance per rack when compared to traditional x86 CPUs. The AGI CPU was co-developed with Meta Platforms to optimize infrastructure for the tech giant’s family of apps and custom silicon.
Arm highlighted additional deals with companies including Cerebras, Cloudflare, F5, OpenAI, Positron, Rebellions, SAP, and SK Telecom. Companies are expected to deploy Arm chips for Agentic CPU use cases, including accelerator management, control plane processing, and cloud and enterprise-based API, task, and application hosting.
TSMC is manufacturing the chip using its 3nm process technology. Arm is also partnering with OEMs and ODMs, including ASRock Rack, Lenovo, Quanta Computer, and Supermicro, with early systems currently available, as the broader availability is due in the second half of the year.