Samsung to begin mass production of HBM4 chips in February
Samsung Electronics will begin mass production of sixth-generation HBM4 memory chips in February 2026 at its Pyeongtaek campus in South Korea, with rival SK Hynix also starting production around the same time.
The chips will be installed in Nvidia's Vera Rubin AI accelerator system launching in Q3 2026 and Google's Tensor Processing Units, following Samsung's successful quality tests with Nvidia that showed speeds up to 11.7Gbps using advanced 10nm-class fabrication.
The production launch addresses a critical industry-wide memory shortage affecting major AI firms including Microsoft, Amazon, and OpenAI, with HBM capacity already sold out through 2026 and analysts projecting Samsung's memory profits could grow 310% in 2026.
Read more
Samsung Electronics will begin mass production of sixth-generation HBM4 memory chips in February 2026 at its Pyeongtaek campus in South Korea, with rival SK Hynix also starting production around the same time.
The chips will be installed in Nvidia's Vera Rubin AI accelerator system launching in Q3 2026 and Google's Tensor Processing Units, following Samsung's successful quality tests with Nvidia that showed speeds up to 11.7Gbps using advanced 10nm-class fabrication.
The production launch addresses a critical industry-wide memory shortage affecting major AI firms including Microsoft, Amazon, and OpenAI, with HBM capacity already sold out through 2026 and analysts projecting Samsung's memory profits could grow 310% in 2026.
Read more
