Intel says select Xeon 6+ rack configurations can scale beyond 46,000 CPU cores while drawing 133.45 kW at the rack level, giving agentic-AI systems a dense x86 layer for orchestration, data movement and tool execution.
The design was shown at OCP Global Summit 2026 alongside Intel’s GPU D-series. Many of the Xeon systems follow Open Rack v3, which matters because operators can combine high-density CPUs with accelerators while keeping the surrounding rack, power and management architecture modular.
Intel’s D-series adds the accelerator side of the stack: up to 480 GB of LPDDR5X in partner designs, 1.5 TB/s memory bandwidth and 1.3 PFLOPS of FP8 compute within a 350 W envelope. Quartz noted that a single card can support models up to 300 billion parameters, while an eight-GPU system can move beyond one trillion parameters.
The rack announcement builds on BitcoinVersus.Tech coverage of Intel’s 480 GB AI GPU, the Open Compute Project’s AI engineering work and Cisco’s 51.2 Tb/s AI-rack silicon. The larger trend is clear: AI infrastructure is being optimized at rack scale, not component by component.
BitcoinVersus.Tech Editor’s Note:
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. If you would like to support to help further secure the integrity of our research initiatives, please donate using the QR code provided
For best QR-code scanning results, please scan this donation code from a desktop or larger screen. Mobile screens may prevent camera recognition.
Leave a Reply