SK hynix finishes HBM4 development, ready for mass production: 10Gbps per pin, above 8Gbps spec

SK hynix has confirmed it has successfully completed the development and finished preparation of its next-gen HBM4 memory, ready for ultra-high-performance AI, and will enter mass production for the world's first time.

2

The South Korean firm has said that it's successfully completed development and based on this technological achievement, SK hynix has prepared HBM4 mass production to "lead the AI era". We will see SK hynix HBM4 memory inside of next-gen AI chips like NVIDIA's upcoming Rubin AI GPUs in 2026.

Joohwan Cho, Head of HBM Development at SK hynix, who has led the development, said: "Completion of HBM4 development will be a new milestone for the industry. By supplying the product that meets customer needs in performance, power efficiency and reliability in a timely manner, the company will fulfill time to market and maintain competitive position".

On a new post on its website, SK hynix says that with the recent dramatic increase in AI demand and data processing, the needs for high bandwidth memory (HBM) for faster system speed are "surging". Not only that, but securing memory power efficiency has emerged as a key requirement as power consumption for data center operations has increased.

SK hynix has far exceeded the JEDEC standard operating speed of 8Gbps for HBM4, pushing up into 10Gbps operating speeds for its HBM4 memory. HBM4 has the industry's best data processing speed and power efficiency, doubling the bandwidth through the adoption of 2048 I/O terminals over the previous-gen HBM3 standard, and power efficiency has climbed by over 40%.

SK hynix expects AI service performance improvements by up to 69% when HBM4 is used, which will "lead to solving data bottlenecks and significantly reduce data center power costs".