Samsung begins mass production of first commercial HBM4 for artificial intelligence computing

Samsung has started mass production and customer shipments of its industry-first HBM4 memory, manufactured on an advanced 6th-generation 10 nm-class DRAM process. The company is positioning the new high bandwidth memory as a performance driver for next-generation artificial intelligence workloads.

Samsung Electronics announced that it has begun mass production of its HBM4 memory and has already shipped commercial products to customers, becoming the first company in the industry to reach this milestone. The launch secures an early leadership position for Samsung in the HBM4 market, which targets high performance computing and artificial intelligence workloads that demand significantly higher memory bandwidth and efficiency.

To achieve this, Samsung is using its most advanced 6th-generation 10 nanometer (nm)-class DRAM process (1c), which it says delivers stable yields and industry leading performance from the very start of mass production, and this was accomplished without any additional redesigns. The company emphasizes that this proactive move to the latest process node is intended to underpin both performance and manufacturability as demand for high bandwidth memory accelerates in data centers and artificial intelligence infrastructure.

Samsung departed from the conventional approach of relying on existing proven designs and instead adopted leading edge technologies such as the 1c DRAM node and a 4 nm logic process for HBM4. According to Sang Joon Hwang, Executive Vice President and Head of Memory Development at Samsung Electronics, leveraging process competitiveness and design optimization provides substantial performance headroom. The company positions this headroom as key to meeting customers’ escalating requirements for higher performance in artificial intelligence computing when they need it.

70

Impact Score

Nvidia DGX Spark brings desktop supercomputing to universities worldwide

Nvidia’s DGX Spark desktop supercomputer is giving universities petaflop-class Artificial Intelligence performance at the lab bench, supporting projects from neutrino astronomy at the South Pole to radiology report analysis and robotics on campus. Institutions are using the compact systems to run large models locally, protect sensitive data and prototype workflows before scaling to big clusters or cloud resources.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.