NVIDIA and SK Hynix reach long-term cooperation to jointly develop next-generation AI memory

[ Gearbest Technology News]On June 8, Nvidia and SK Hynix announced the establishment of a multi-year technology partnership. The two parties will conduct in-depth cooperation around the demand for high-performance memory in the artificial intelligence era, jointly develop advanced memory products for the next generation of AI computing platforms, and further promote the application of AI technology in semiconductor design and manufacturing.

NVIDIA

According to information announced by both parties, this cooperation will cover multiple important computing platforms. SK Hynix will jointly develop special memory products with NVIDIA, which will be used in NVIDIA Vera Rubin AI supercomputer, NVIDIA Vera CPU, NVIDIA RTX Spark PC and NVIDIA Jetson Thor robot computing platform and other product systems. Through this cooperation, SK Hynix will further participate in emerging markets such as AI infrastructure, personal AI, and physical AI that NVIDIA is laying out.

Both parties stated that as the construction speed of global AI factories continues to accelerate, the demand for high-performance memory also continues to grow. Due to the long development cycle of advanced memory products, this multi-year cooperation agreement will help both parties plan technology routes and production capacity layout in advance to ensure that future products can keep pace with the development pace of Nvidia's AI infrastructure.

In addition to memory technology research and development, the two parties also plan to conduct more in-depth cooperation in the field of semiconductor manufacturing. According to the plan, SK Hynix will introduce NVIDIA's CUDA-X library and PhysicsNeMo framework to improve the efficiency of key links such as chip simulation and lithography calculations, thereby accelerating the chip design and manufacturing process.

At the same time, SK Hynix will also use NVIDIA Omniverse and cuOpt technologies to build a digital twin system for the wafer factory to digitally simulate and optimize the production process, further improve factory operation efficiency, and promote the development of manufacturing links to a higher degree of automation and intelligence.