Nvidia shifts HBM4 mix toward 8-high as memory supply stays tight
DIGITIMESen

Nvidia has reportedly asked Samsung Electronics and SK Hynix to increase the share of 8-high HBM4 supplied for its Vera Rubin accelerator platforms, as the chipmaker seeks greater flexibility in balancing thermal constraints, packaging yields and memory availability amid tight supply.
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
- AI 資安成新戰場 CrowdStrike 財報報喜盤後噴近 9%TechNews (TW) · August 27, 2026
- OpenAI 宣布 ChatGPT 迎來重大升級,無須碰觸帳密也能幫你完成網站代辦任務TechNews (TW) · August 27, 2026
- OpenAI says reward hacking, an AI alignment problem in which a model takes unintended actions to achieve a goal, was a primary driver of the Hugging Face breach (Hayden Field/The Verge)Techmeme · August 27, 2026
- DBS, Stripe team up on AI, cross-border payments in AsiaTech in Asia · August 27, 2026
- SEC filing: OpenAI invests ~$400M in its second startup fund, after it raised $175M for its first fund in 2021 from outside investors, including Microsoft (Sarah Klearman/Wall Street Journal)Techmeme · August 27, 2026
- Intelligence artificielle : Nvidia cherche à rassurer les marchés fébriles sur la croissance de la demandeLe Monde Pixels · August 27, 2026