AMD Unveils World's First 7nm Datacenter GPUs

New ultra-fast accelerators will support deep learning and other HPC scenarios

AMD has announced the Radeon Instinct MI60 and MI50 accelerators, which the company calls “the world's first 7nm datacenter GPUs.” The new GPUs are intended for deep learning, cloud computing, and rendering applications.

According to the press release, the MI60 and MI50 “provide ultra-fast floating-point performance and hyper-fast HBM2 (second-generation High-Bandwidth Memory) with up to 1 TB/s memory bandwidth speeds. They are also the first GPUs capable of supporting next-generation PCIe® 4.0 interconnect, which is up to 2X faster than other x86 CPU-to-GPU interconnect technologies, and feature AMD Infinity Fabric™ Link GPU interconnect technology that enables GPU-to-GPU communications that are up to 6X faster than PCIe® Gen 3 interconnect speeds.”

AMD also announced a new version of its ROCm development platform that includes support for the high-end acceleration features of the MI60 and MI50 GPUs.