Intel Xeon 6 used as Host CPUs in NVIDIA DGX Rubin NVL8 Systems

What's New: Today at NVIDIA GTC 2026, Intel announced that Intel Xeon 6 is being used as the processor for NVIDIA DGX Rubin NVL8 systems. This highlights Xeon's role in providing architectural continuity and scalability for GPU-accelerated AI systems as workloads shift toward massive, real-time inference.

"AI is shifting from large-scale training to real-time, everywhere inference-driven by agentic AI and reasoning systems," said Jeff McVeigh, corporate vice president and general manager, Data Center Strategic Programs at Intel. "In this new era, the host CPU is mission-critical. It governs orchestration, memory access, model security, and throughput across GPU-accelerated systems. Intel Xeon 6 delivers leadership performance, efficiency, and compatibility with the extensive x86 software ecosystem that customers rely on to scale inference workloads."

Why It Matters: As organizations continue to deploy AI systems, inference is increasingly defined not only by GPU throughput but also by CPU-led system performance, with the host CPU shaping overall cluster efficiency and total cost of ownership. It is also responsible for critical functions such as memory management, task orchestration, and workload distribution, while ensuring the security, reliability, and operational continuity essential to modern AI infrastructure.

Building on these system-level requirements, Intel Xeon processors are used as the host CPU for DGX Rubin NVL8 systems due to their capability to support fast memory speeds, balanced performance across a range of workloads, lower long-term total cost of ownership (TCO), and their mature, enterprise-proven software ecosystem. Additionally, Intel's robust PCIe and I/O capabilities further strengthen Xeon's role as a high-bandwidth, low-latency platform across diverse workloads.

• Efficient performance per watt
• Optimized support throughout the ecosystem AI software stack, including new support for NVIDIA Dynamo to enable heterogeneous inference across CPU and forthcoming GPUs
• Proven reliability across mission critical environments
• Superior orchestration of GPU accelerated, heterogeneous systems

This selection reinforces Intel Xeon as a cornerstone of modern AI infrastructure, enabling scalable deployment across modern data centers, the cloud and edge use cases. As AI inference scales, end-to-end confidential computing becomes essential - from CPU to GPU data paths. Intel Trust Domain Extensions (TDX) adds hardware-based isolation and attestation further reinforcing the selection of Xeon as the secure foundation for modern AI clusters.

About the new Intel and NVIDIA collaboration: NVIDIA DGX Rubin NVL8 systems integrate Intel Xeon 6 processors, building on the architectural foundation established with Intel Xeon 6776P in current NVIDIA Blackwell-based platforms, including DGX B300 systems. By building on this proven foundation, Intel is helping to carry forward the performance, experience, and system-level expertise into the new DGX Rubin NVL8 systems.

Intel engineered Xeon to help these systems get the most out of their GPUs, using features like Priority Core Turbo to keep data flowing to GPUs - and with strong single-thread performance handling orchestration, scheduling, and data movement, Xeon helps ensure everything runs smoothly and efficiently even as inference workloads grow more complex.

Key features of Intel Xeon 6 include:

• Up to 8 TB system memory to support large models and growing KV caches
• 3X higher memory bandwidth gen-on-gen with MRDIMM technology¹, improving data feed rates to GPUs
• Industry-leading PCIe 5.0 lanes to support AI accelerators and other devices
• Confidential computing across CPU-GPU data paths with Encrypted Bounce Buffer
• Hardware-rooted isolation safeguards AI data and models while in use

What's Next: Learn more at the Intel booth #3100 on the show floor in the San Jose Convention Center.