SK hynix says its ultra-next-gen HBM4E in 2026, ready for the world of next-gen AI GPUs

SK hynix has announced it plans to complete the development of its next-gen HBM4E memory by as early as 2026, preparing for the next-gen AI GPUs of the future.

2

SK hynix's head of the HBM advanced technology team, Kim Gwi-wook, announced the news this week of the direction of next-generation HBM development at the International Memory Workshop (IMW 2024). HBM was developed by SK hynix in 2014, with HBM2 (2nd generation) in 2018, HBM2E (3rd generation) in 2020, HBM3 (4th generation) in 2022, and HBM3E (5th generation) was introduced this year.

There's a two-year cadence between HBM generations, with HBM3E unleashed this year, it means that HBM4 (6th generation) should drop in 2025, and HBM4E (7th generation) in 2026. That's a bit faster than two years for HBM4 and HBM4E, which is because SK hynix is predicting that HBM performance advancements would become faster than previous generations.

• Read more: SK hynix says most of its HBM for 2025 is sold out already, 16-Hi HBM4 coming in 2028
• Read more: SK hynix expects HBM memory chip revenues of over $10 billion in 2024
• Read more: SK hynix plans $14.6 billion on new chip fab in South Korea to meet 'soaring demand' of HBM
• Read more: SK hynix and TSMC to work together with HBM4 and next-gen semiconductor packaging technology

Kim said: "Since the first generation of HBM was developed, generations have been developed every two years, but from HBM3E, generations are changing every year".

SK hynix teases that HBM4 will have 1.4x more bandwidth than HBM3E, integration will be 1.3x higher, and we can expect 30% better power efficiency. SK hynix will use existing MR-MUF technology until the mass production of HBM4, while the South Korean giant is also looking into hybrid bonding technology that bonds the top and bottom of DRAM to copper.