Intel Expands Xeon 6 Lineup to Power Next-Generation GPU-Accelerated AI Systems
Santa Clara, California — Intel today announced the expansion of its Intel® Xeon® 6 processor family with three new models purpose-built to support AI systems driven by advanced GPUs. These new CPUs, equipped with high-performance P-cores, integrate Intel’s innovative Priority Core Turbo (PCT) technology and Intel® Speed Select Technology – Turbo Frequency (Intel® SST-TF). This combination allows for dynamic, customizable core frequency scaling—optimized to boost GPU performance under demanding AI workloads.
One of the new Xeon 6 processors, the Intel® Xeon® 6776P, is already shipping as the host CPU in the newly launched NVIDIA DGX B300, NVIDIA’s next-generation AI acceleration platform. The Xeon 6776P plays a critical role in orchestrating and managing the system’s operations, leveraging its extensive memory capacity and high bandwidth to meet the escalating data requirements of modern AI models.
“These new Xeon SKUs demonstrate the unmatched performance of Xeon 6 processors, making them the ideal CPUs for powering the next generation of GPU-accelerated AI infrastructure,”
said Karin Eibschitz Segal, Intel VP and Acting GM of the Data Center and AI Group.
“We’re excited to strengthen our collaboration with NVIDIA in delivering one of the most powerful AI systems available today.”
Boosting AI Workload Efficiency with Priority Core Turbo
The combination of PCT and Intel SST-TF introduces a significant leap in AI system performance. PCT enables dynamic prioritization of high-performance cores, allowing select cores to operate at elevated turbo frequencies, while others maintain base clock speeds. This approach ensures intelligent CPU resource allocation, ideal for AI tasks that require sequential or serialized processing—feeding data more rapidly to GPUs and maximizing system throughput.
Core Advantages of the New Xeon 6 P-Core Processors
• High Core Counts & Superior Single-Thread Performance:
Each CPU supports up to 128 P-cores, providing balanced workload distribution for compute-intensive AI tasks.
• 30% Faster Memory Bandwidth:
With support for MRDIMM and Compute Express Link (CXL), the new processors deliver leading-edge memory speeds in high-capacity configurations, outperforming competing platforms.
• Expanded I/O Capabilities:
Compared to prior generations, PCIe lane count has been increased by 20%, enabling faster data throughput for I/O-heavy applications.
• Robust Reliability & Serviceability:
The new Xeon processors offer industry-grade reliability, availability, and maintainability (RAS) features—minimizing system downtime and business disruption.
• Intel® Advanced Matrix Extensions (AMX):
Native support for FP16 precision boosts efficiency in data preprocessing and other AI-critical CPU operations.
As AI adoption continues to accelerate, enterprises are upgrading infrastructure to handle increasingly complex workloads. With the ideal blend of performance, efficiency, and scalability, Intel’s Xeon 6 processors with P-cores solidify the company’s leadership in CPU solutions optimized for artificial intelligence across data centers and edge networks.