Micron HBM3E 8-Hi 36GB

The Micron HBM3E 8-Hi 36GB (MT60B4G32HB-48B:A) represents Micron's implementation of the High Bandwidth Memory 3E standard in an eight-layer stacked die configuration, delivering exceptional memory bandwidth and capacity in a compact 2.5D integration footprint. Built on JEDEC HBM3E specifications, this module achieves a per-stack bandwidth of up to 1.2 TB/s, making it purpose-built for compute platforms where memory bandwidth is the primary performance constraint. The 8-Hi stack achieves 36GB capacity per stack through the vertical integration of eight 4.5GB DRAM dies interconnected via Through-Silicon Vias (TSVs), enabling massive parallelism across a 1024-bit wide memory interface per stack.

Designed for integration alongside high-performance processors, GPUs, and AI accelerators via silicon interposer or advanced packaging technologies such as CoWoS (Chip-on-Wafer-on-Substrate), the HBM3E 8-Hi 36GB operates at data rates up to 9.6 Gbps per pin at a nominal supply voltage of 1.2V (VDDQ). Its architecture supports error correction through on-die ECC as well as advanced RAS (Reliability, Availability, and Serviceability) features critical for enterprise and datacenter deployments where data integrity across sustained, high-intensity workloads is non-negotiable. The device is fully compliant with the JEDEC HBM3E standard and is offered in the standard HBM3E physical form factor compatible with leading interposer-based platform designs.

The Micron HBM3E 8-Hi 36GB is specifically engineered for the most demanding AI training and inference pipelines, high-performance computing (HPC) simulations, and memory-bandwidth-intensive datacenter applications. As large language models (LLMs), generative AI frameworks, and scientific computing workloads continue to scale, this product addresses the growing gap between compute throughput and memory subsystem performance. Its combination of high capacity, extreme bandwidth, low latency, and advanced error correction positions it as a foundational memory component for next-generation AI accelerators, HPC nodes, and datacenter-scale infrastructure deployments across enterprise, cloud, and research environments.

Ideal for

• Large-scale AI model training requiring sustained high memory bandwidth for transformer-based and generative AI workloads on GPU and AI accelerator platforms
• High-performance inference serving of large language models (LLMs) where memory capacity per accelerator directly limits achievable batch size and throughput
• High-performance computing (HPC) simulations including computational fluid dynamics, molecular dynamics, and finite element analysis with large in-memory datasets
• Datacenter-scale deep learning research platforms integrating multiple HBM3E stacks via advanced packaging for maximum aggregate bandwidth
• Scientific computing and government/research supercomputing nodes requiring both high capacity and data integrity through on-die ECC under continuous operation
• Next-generation AI accelerator card design for enterprise AI infrastructure, enabling system integrators to build bandwidth-optimized compute modules

Technical specifications

Available from Omnixon Global. Submit an RFQ and our team will confirm configuration and availability for your order.