Samsung Ships Industry-First HBM4E Samples, Pushing AI Memory to 16Gbps Speeds
Samsung Electronics has announced a major breakthrough in AI memory technology, revealing that it has begun shipping the industry's first 12-layer HBM4E (High Bandwidth Memory 4E) samples to major global customers. The development marks a significant leap forward in memory performance for next-generation artificial intelligence workloads.
Building on the successful mass production of its HBM4 memory earlier this year — which was itself an industry first — Samsung's new HBM4E extends the company's lead in the high-bandwidth memory market. The announcement comes at a time when demand for advanced AI infrastructure is exploding, with hyperscale data centers and large language model (LLM) training requiring ever-faster memory solutions.
Next-Level Performance Numbers
Samsung's HBM4E delivers a stable pin speed of 14 gigabits per second (Gbps), scalable up to 16Gbps to support increasingly intensive data processing requirements. This represents a more than 20 percent increase over its predecessor HBM4. The new memory achieves memory bandwidth of up to 3.6 terabytes per second (TB/s) per stack, helping maximize computing performance for large language models and next-generation AI systems.
Each 12-layer HBM4E stack offers 48 gigabytes (GB) of capacity — more than a 30 percent increase over the previous generation. Samsung plans to expand the lineup to include 32GB (8-layer) and 64GB (16-layer) configurations based on customer requirements.
Built on Proven Technology
The HBM4E takes full advantage of Samsung's comprehensive semiconductor capabilities, leveraging the same leading-edge technologies refined through the company's HBM4 production experience. This includes the industry's most advanced 6th-generation 10-nanometer (nm)-class DRAM process (1c) and Samsung Foundry's 4nm logic base die, allowing the HBM4E to secure enhanced process stability and manufacturability.
"Following the successful mass production of HBM4, Samsung has once again demonstrated its distinct technological edge with HBM4E," said Sang Joon Hwang, Executive Vice President and Head of Memory Development at Samsung Electronics. "Through our advanced manufacturing capabilities and preemptive infrastructure investments, we will continue to drive the growth of the global AI memory market."
AI Memory Market Context
The high-bandwidth memory market has become one of the most competitive battlegrounds in the semiconductor industry, driven by insatiable demand from AI companies building ever-larger neural networks. Samsung's HBM4, introduced in February, was the first in the industry to enter mass production and has set the bar with speeds of 11.7Gbps in system-in-package (SiP) tests. Customer feedback has been highly positive, particularly regarding its performance and energy efficiency.
With the new HBM4E samples now shipping, Samsung is positioning itself to maintain its leadership as the AI memory market continues its explosive growth. The company plans to begin mass production for HBM4E aligned with customer schedules, following initial sample shipments and optimization.
What This Means for AI
Higher memory bandwidth directly translates to faster AI training and inference times. For companies building large language models and other AI systems, Samsung's HBM4E could significantly reduce the time and energy required to train cutting-edge models. The combination of higher speeds, greater capacity, and improved energy efficiency makes HBM4E a critical component for the next generation of AI infrastructure.
With its comprehensive portfolio spanning memory, foundry, logic design, and advanced packaging, Samsung is positioned to ensure stable semiconductor supply for the booming AI market, helping accelerate innovation across the entire AI ecosystem.
Image Source: Samsung
