Three of Intel’s recently unveiled Xeon 6 CPUs will help host and orchestrate AI workloads in Nvidia’s Blackwell-based DGX B300 systems, marking a strategic win for the chipmaker as it fights to reclaim relevance in AI infrastructure.
The chips feature Intel’s new Priority Core Turbo (PCT) tech, which dynamically pushes “high-priority cores” to higher turbo frequencies, helping the CPU feed data to GPUs faster, crucial for keeping those expensive GPUs fully utilised.
Intel claims the CPUs offer 30% faster memory speeds than rival hardware, enhanced I/O throughput, and up to 128 cores, making them well-suited for inference and orchestration tasks in GPU-heavy AI environments.
“These new Xeon SKUs demonstrate the unmatched performance of Intel Xeon 6, making it the ideal CPU for next-gen GPU-accelerated AI systems,” said Karin Eibschitz-Segal, corporate VP and interim general manager of the Data Centre Group at Intel.
Since taking over from Pat Gelsinger late last year, CEO Lip-Bu Tan’s turnaround plans have focused on doubling down on Intel’s historically core markets: PCs and data centres.
The team up with Nvidia for the DGX B300s marks a much-needed win for Intel, as it looks to make up market share that has been steadily eroded by AMD’s EPYC line and the growing competition from Arm.
While the DGX B300s are still very much Nvidia’s show, powered by Blackwell GPUs and designed for the most demanding AI training workloads, the inclusion of Intel silicon in such a flagship system offers a strong signal to the market.
The Xeon 6 family, and particularly the 6776P variant chosen for DGX, offers a blend of high core count, memory bandwidth, and I/O throughput that is increasingly important in AI workflows, especially when models and datasets are too large to manage purely within GPU memory.
It also shows Intel’s willingness to work closely with ecosystem leaders like Nvidia, despite the fact that the pair compete in adjacent areas, in a bid to secure relevance amid a once-in-a-generation compute shift.
“We’re thrilled to deepen our collaboration with Nvidia to deliver one of the industry’s highest-performing AI systems, helping accelerate AI adoption across industries,” Eibschitz-Segal added.
RELATED STORIES
Intel launches new Xeon 6 lineup targeting data centre and networking
Intel eyes sale of network and edge unit as cost-cutting continues
Intel bets on 18A and chiplets in bid to revive its foundry fortunes
