Samsung, SK Hynix aim to ease AI memory bottlenecks with zHBM, HBF
DIGITIMESen

Samsung Electronics and SK Hynix are developing new memory technologies to address growing bottlenecks in AI systems, with Samsung exploring a design that stacks high-bandwidth memory (HBM) directly above graphics processors and SK Hynix advancing high-bandwidth flash (HBF) to expand memory capacity for AI inference.
This is a short summary published by AI Global Wire. The full article is owned and hosted by DIGITIMES — open it there to read it in full.
Read the full story at DIGITIMESRelated AI news
- Schneider Electric lifts 2026 outlook on strong AI demandTech in Asia · October 16, 2026
- K�nstliche Intelligenz: Google streicht Gemini Pro aus dem AI-Plus-AboGolem.de · October 16, 2026
- AstraZeneca strikes $1.5b AI drug discovery deal with InsitroTech in Asia · October 16, 2026
- Kuaishou shares rise 4% as Kling AI launches video modelTech in Asia · October 16, 2026
- »De involverede skal stå skoleret«: Nu starter stor undersøgelse af »ulovligt« AI-forsøgIngeniøren · October 16, 2026
- Bringing operational AI to legacy manufacturingTech in Asia · October 16, 2026