Intel details D-series inference GPUs as customer sampling proceeds

Intel disclosed new specifications for its D-series inference GPUs and introduced Xeon-based rack designs at the Open Compute Project Global Summit, with customer sampling of the GPUs underway. The company says the GPU design combines up to 480 GB of LPDDR5X memory, 1.5 TB/s of memory bandwidth and 1.3 petaflops of FP8 compute within a 350-watt power envelope. Its standard PCIe format and air cooling are intended to fit existing data centers. Intel's own card has 160GB of memory; partner designs can offer up to 480GB. Select Xeon 6+ configurations exceed 46,000 cores per rack at 133.45 kW total rack power. Intel cautions that performance varies by configuration and use.