Samsung Begins Mass Production of HBM4 Memory Chips This Month

Samsung starts mass-producing advanced HBM4 memory chips this month for NVIDIA's upcoming AI accelerators, boosting performance for large-scale computing.

Feb 10, 2026
4 min read
Set Technobezz as preferred source in Google News
Technobezz
Samsung Begins Mass Production of HBM4 Memory Chips This Month

Don't Miss the Good Stuff

Get tech news that matters delivered weekly. Join 50,000+ readers.

Samsung will begin mass production of sixth-generation HBM4 memory chips this month, according to multiple industry reports. The South Korean tech giant plans to ship the advanced semiconductors to NVIDIA by mid-February for use in upcoming Vera Rubin AI accelerators.

Samsung reportedly passed NVIDIA's final verification tests and secured purchase orders for the HBM4 chips. Production timing matches NVIDIA's launch plans for its next-generation AI accelerator, known internally as Vera Rubin.

The company's stock gained between 4.9% and 6.4% on Monday following the production announcement. NVIDIA shares jumped 7.87% on the news, while competitor SK Hynix saw a 5.72% increase. Micron Technology stock fell 3.2% as Samsung's accelerated timeline creates competitive pressure.

Samsung's HBM4 chips feature data processing speeds up to 11.7 Gbps, exceeding industry standards and offering 22% higher performance than previous generation HBM3E chips. The high-bandwidth memory is designed specifically for AI servers and large-scale computing applications.

Micron plans to ramp up its own HBM4 production during the second quarter of 2026, putting the company roughly one quarter behind Samsung's schedule. Micron CEO Sanjay Mehrotra outlined the company's HBM4 plans during its most recent earnings call.

Samsung's market capitalization sits at $694.62 billion following the stock surge. The company reported revenue of $223.32 billion with a 7% growth rate over three years.

Analysts note that AI accelerator vendors are moving toward three-supplier sourcing strategies, creating space for Samsung, SK Hynix, and Micron. Companies previously relied on just two suppliers for their high-bandwidth memory needs.

Samsung maintains its position as the only semiconductor manufacturer capable of providing solutions across logic, memory, foundry, and packaging. The company plans to use this integrated capability to secure additional AI chip customers beyond NVIDIA.

The HBM4 production represents a key win for Samsung in the AI supply chain race, with the company's chips offering 22% higher performance than previous generation HBM3E memory.

Share this article

Help others discover this content

More in News