Intel has announced the availability of its Xeon 6+ processors at Computex 2026, introducing a new generation of data centre CPUs built on the company’s Intel 18A process technology. The processors are designed for cloud-native applications, AI inference and network-intensive workloads, as Intel seeks to strengthen the role of CPUs in increasingly AI-driven data centres.
According to Intel, the rise of inference and agentic AI workloads is shifting computing requirements within modern data centres. While AI training deployments have traditionally relied heavily on accelerators, the company believes the growing demand for orchestration, concurrency and data movement is increasing the importance of high-performance CPUs.
The new Xeon 6+ processors are engineered to address these requirements through higher performance density, improved power efficiency and support for large-scale deployment scenarios. Intel says the processors are optimised for workloads where throughput, latency consistency and rack-level efficiency are key considerations.
Built on Intel 18A, Xeon 6+ marks the first use of the manufacturing node in a data centre processor. Intel claims the architecture is capable of supporting AI-focused rackscale deployments, with a single liquid-cooled rack configuration accommodating up to 36,864 CPU cores across 32U of compute space. The company positions the platform as a solution for organisations deploying large numbers of AI agents and inference workloads at scale.
Alongside the processor launch, Intel highlighted ongoing collaborations with partners including Foxconn and SambaNova to develop rackscale AI infrastructure. The company also showcased deployments that combine Xeon processors with dedicated AI accelerators to improve inference performance while maintaining power efficiency.
The launch comes as Intel continues to expand its AI portfolio beyond traditional CPUs, with the company increasingly focusing on infrastructure designed for inference and agentic AI applications. Xeon 6+ is expected to play a central role in those efforts as enterprises look to deploy AI workloads at greater scale over the coming years.
