Arm Holdings has made a historic shift by designing and selling its own data center chips for the first time in the company's 30-year history. The new Arm AGI CPU is purpose-built for artificial intelligence infrastructure, specifically for handling agentic AI workloads where AI systems continuously reason, plan and act. This move represents a fundamental change in how Arm operates, expanding beyond licensing chip designs to manufacturing finished silicon products .

Why Is Arm Suddenly Building Its Own Chips?

For decades, Arm made money by licensing its chip architecture to other companies like Qualcomm, Apple and Samsung, who then manufactured their own products. But the rise of agentic AI is reshaping data center demands in ways that existing chip designs cannot efficiently handle. As AI agents become more prevalent, data centers need significantly more CPU capacity to manage reasoning, coordination and data movement between systems. Organizations scaling agent-driven applications are expected to require more than 4x the current CPU capacity per gigawatt of power, all while operating within strict power budgets .

Arm's CEO Rene Haas explained the strategic reasoning behind this unprecedented move. "AI has fundamentally redefined how computing is built and deployed," he stated. "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 global scale."

Rene Haas

What Makes the Arm AGI CPU Different From Existing Processors?

The Arm AGI CPU is engineered specifically for AI-scale infrastructure without the overhead and complexity of traditional x86 processors used in most data centers. The chip delivers measurable advantages across three critical dimensions :

• Performance Capability: Up to 136 Arm Neoverse V3 cores per CPU with 6 gigabytes per second of memory bandwidth per core at sub-100 nanosecond latency, enabling leading performance per core, system-on-chip, blade and rack.
• Power Efficiency: Operates at a 300-watt thermal design power with a dedicated core per program thread, eliminating throttling and idle threads that waste energy in traditional processors.
• Density and Cooling: Supports high-density 1U server chassis with air-cooled deployments reaching 8,160 cores per rack, or liquid-cooled systems delivering over 45,000 cores per rack.

These specifications translate into concrete business benefits. The Arm AGI CPU delivers more than 2x performance per rack compared with x86 CPUs, potentially enabling up to $10 billion in capital expenditure savings per gigawatt of AI data center capacity . For context, a gigawatt represents the power consumption of a large data center facility, so these savings apply to enterprise-scale deployments.

How to Evaluate Whether Arm AGI CPUs Make Sense for Your Organization

If your company operates or plans to build AI infrastructure, consider these practical factors when evaluating the Arm AGI CPU :

• Workload Type: Assess whether your primary use cases involve agentic AI, accelerator management, control plane processing, or cloud-based API and application hosting, where the Arm AGI CPU is optimized.
• Power Budget Constraints: Calculate your facility's power limitations and compare whether the 300-watt TDP and high core density per rack align with your infrastructure goals and cooling capabilities.
• Capital Expenditure Timeline: Early systems are available now from partners like Lenovo, Supermicro, ASRock Rack and Quanta Computer, with broader availability expected in the second half of 2026.
• Ecosystem Support: Verify that your software stack, middleware and applications can run on Arm architecture, as over 50 leading companies including AWS, Google, Microsoft, NVIDIA and TSMC are supporting this platform expansion.

Meta, the social media and AI giant, served as the lead partner and co-developer of the Arm AGI CPU. Santosh Janardhan, head of infrastructure at Meta, explained the company's rationale: "Delivering AI experiences at global scale demands a robust and adaptable portfolio of custom silicon solutions, purpose-built to accelerate AI workloads and optimize performance across Meta's platforms. We worked alongside Arm to develop the Arm AGI CPU to deploy an efficient compute platform that significantly improves our data center performance density and supports a multi-generation roadmap for our evolving AI systems."

Beyond Meta, Arm has secured commitments from a diverse set of customers including Cerebras, Cloudflare, F5, OpenAI, Positron, Rebellions, SAP and SK Telecom. These organizations plan to deploy the Arm AGI CPU for critical use cases such as accelerator management, control plane processing and cloud-based API hosting .

What Does This Mean for the Broader AI Chip Market?

Arm's entry into finished silicon products signals a major shift in how the AI infrastructure market is evolving. For years, Nvidia dominated AI chip sales with its graphics processing units (GPUs), while companies like Intel and AMD controlled the CPU market. By designing its own data center CPUs, Arm is positioning itself to capture a significant portion of the CPU workloads that support AI systems, rather than relying solely on licensing fees from other manufacturers .

The ecosystem support is remarkably broad. More than 50 leading companies across hyperscale cloud providers, silicon manufacturers, memory makers, networking vendors, software companies and system designers have publicly committed to supporting the Arm AGI CPU platform. This includes industry giants such as AWS, Broadcom, Google, Marvell, Micron, Microsoft, NVIDIA, Samsung, SK hynix and TSMC .

The timing of this launch reflects a fundamental recognition across the industry that agentic AI represents a new computing paradigm. Unlike traditional AI model training, which is dominated by GPU workloads, agentic systems require continuous CPU processing to manage reasoning loops, coordinate between multiple AI agents and handle data movement. This creates demand for a new class of processors optimized for these specific tasks, and Arm is positioning the AGI CPU as the foundational platform for this emerging infrastructure category.