Micron announces HBM3e memory enters volume production to supply NVIDIA's H200 AI GPU.

Micron Technology has announced the start of volume production of its HBM3e memory, with its known good stack dies (KGSDs) shipping as part of NVIDIA's upcoming H200 AI GPU. This marks a significant milestone in delivering advanced memory solutions to support the growing demands of AI workloads.

The NVIDIA H200 AI GPU will be equipped with up to 141GB of Micron's high-speed HBM3e memory, sourced from its mass-produced 24GB 8-Hi HBM3e memory modules. These modules deliver data transfer rates of 9.2GT/s with a peak memory bandwidth exceeding 1.2TB/sec per GPU, representing a 44% increase in memory bandwidth over the previous HBM3 generation. In total, the H200 will feature up to 4.8TB/sec of memory bandwidth, substantially improving upon the H100 AI GPU, which contains 80GB of HBM3 memory and up to 3.35TB/sec of memory bandwidth.

Sumit Sadana, executive vice president and chief business officer at Micron Technology, stated: "Micron is delivering a trifecta with this HBM3E milestone: time-to-market leadership, best-in-class industry performance, and a differentiated power efficiency profile." Sadana further emphasized that "AI workloads are heavily reliant on memory bandwidth and capacity, and Micron is very well-positioned to support the significant AI growth ahead through our industry-leading HBM3E and HBM4 roadmap, as well as our full portfolio of DRAM and NAND solutions for AI applications." The expanded memory capacity and bandwidth of the H200 directly address the critical need for memory solutions to keep pace with the expanding demands of AI workloads as demand continues to surge.