Intel’s Crescent Island AI Server GPU Gets a Name and Specs: Meet the Intel GPU D-Series

Intel has unveiled the Intel GPU D-Series, the official branding for its Crescent Island data center AI inference cards. The D270 model delivers 1.3 PetaFLOPS of dense FP8 performance and 1.5TB/second memory bandwidth. It features 32 Xe3P cores, 256 XMX engines, and 32 MB of unified L2 cache. Supporting formats from FP4 to FP64, the GPU includes up to 480 GB of LPDDR5X memory and a media engine with four encoders and four decoders.