Arm’s newly introduced AGI CPU marks a notable shift in how computing hardware is designed for artificial intelligence. Unlike traditional processors built for general-purpose workloads, this chip is engineered specifically for “agentic AI,” systems capable of reasoning, planning, and acting with minimal human oversight, tells Live Science.
The processor is based on Arm’s latest architecture and built using advanced 3-nanometer manufacturing technology. It can integrate up to 136 high-performance cores running at speeds approaching 3.7 GHz. Each core is paired with significant memory bandwidth, enabling the chip to handle the complex, parallel workloads required by AI agents operating in real time.
What distinguishes this CPU is its specialization. Rather than supporting a wide range of legacy software, it strips away unnecessary compatibility layers to focus entirely on AI inference and orchestration tasks. This streamlined design allows it to deliver more than double the performance per server rack compared to conventional x86 processors, while also maintaining energy efficiency, an increasingly critical factor in large-scale data centers.
The AGI CPU is also designed for scalability. When deployed in server configurations, thousands of cores can be combined into a single rack, enabling massive parallel processing. This is particularly important for agentic AI systems, which rely heavily on CPUs for coordination, memory handling, and decision-making rather than just raw numerical computation handled by GPUs.
Industry partners, including major AI and cloud companies, are already aligning with the platform, signaling confidence in its potential. The chip represents Arm’s move beyond licensing designs to delivering full silicon solutions, positioning it as a direct competitor in the data-center hardware space.
Overall, the AGI CPU reflects a broader transition in AI infrastructure. As artificial intelligence evolves from passive models to autonomous agents, the role of CPUs is expanding, becoming central to managing the logic, coordination, and continuous operation that define the next generation of intelligent systems.