Home
News
Products
Corporate
Contact
 
Sunday, June 15, 2025

News
Industry News
Publications
CST News
Help/Support
Software
Tester FAQs
Industry News

Micron Technology starts shipping HBM4 samples to key customers


Thursday, June 12, 2025

This is being described as a significant milestone for the company and extends Micron’s leadership in memory performance and power efficiency for AI applications.

Built on its established 1ß (1-beta) DRAM process, 12-high advanced packaging technology and its memory built-in self-test (MBIST) feature, Micron’s HBM4 has been developed to provide seamless integration for those developing next-generation AI platforms.

At a time when the importance of high-performance memory has never been greater, fuelled by its critical role in supporting the growing demands of AI training and inference workloads in data centres, the ability to effectively manage inference becomes more important.

Micron HBM4 features a 2048-bit interface, achieving speeds greater than 2.0 TB/s per memory stack and more than 60% better performance over the previous generation. This expanded interface enables rapid communication and a high-throughput design that accelerates the inference performance of large language models and chain-of-thought reasoning systems.

In addition, the HBM4 features over 20% better power efficiency compared to Micron’s previous-generation HBM3E products, and this improvement provides maximum throughput with the lowest power consumption to maximise data centre efficiency.

Generative AI use cases continue to increase, and HBM4 is a crucial enabler, driving quicker insights and discoveries in areas such as healthcare, finance and transportation.

Commenting Raj Narasimhan, senior vice president and general manager of Micron’s Cloud Memory Business Unit said, "Building on the remarkable milestones achieved with our HBM3E deployment, we continue to drive innovation with HBM4 and our robust portfolio of AI memory and storage solutions. Our HBM4 production milestones are aligned with our customers’ next-generation AI platform readiness to ensure seamless integration and volume ramp."

Micron said that it plans to ramp HBM4 production in calendar year 2026 and will be in alignment with their customers’ next-generation AI platforms.

By: DocMemory
Copyright © 2023 CST, Inc. All Rights Reserved

CST Inc. Memory Tester DDR Tester
Copyright © 1994 - 2023 CST, Inc. All Rights Reserved