NVIDIA and SK Hynix Announce Multi-Year AI Memory Partnership
The partnership ties SK Hynix's memory production directly to NVIDIA's next-generation product lineup across data centers, AI PCs, and robotics, signaling a strategic move to secure supply for a shortage that Huang expects to last years.
Key Facts
- NVIDIA and SK Hynix announced a multi-year technology partnership on June 7 to jointly develop next-generation memory for AI data centers.
- The partnership covers memory for NVIDIA's Vera Rubin AI supercomputer, Vera CPU, RTX Spark-equipped PCs, and Jetson Thor robotics platforms.
- SK Hynix will expand development and supply of advanced memory tailored to NVIDIA's next-generation product lineup.
- NVIDIA CEO Jensen Huang said AI memory shortages could persist for several years across areas from wafers to silicon photonics, according to Reuters.
Reporting from 2 sources: ASCII.jp, GIGAZINE.
NVIDIA and SK Hynix announced a multi-year technology partnership on June 7 to jointly develop next-generation memory for AI data centers, covering the Vera Rubin AI supercomputer, Vera CPU, RTX Spark-equipped PCs, and Jetson Thor robotics platforms. SK Hynix will expand development and supply of advanced memory to address chronic shortages that NVIDIA CEO Jensen Huang said could persist for years.
NVIDIA and SK Hynix announced a multi-year technology partnership on June 7 to jointly develop next-generation memory for AI data centers, which the companies refer to as AI Factories. The agreement covers memory for NVIDIA's Vera Rubin AI supercomputer, Vera CPU, RTX Spark-equipped PCs, and Jetson Thor robotics computing platforms. SK Hynix will produce memory tailored to NVIDIA's next-generation product lineup across data centers, AI PCs, and robots. The deal comes as AI memory faces chronic shortages. Reuters reported on June 7 that NVIDIA CEO Jensen Huang indicated shortages could persist for several years across areas from wafers to silicon photonics.
Synthesized by Yomimono from the 2 cited sources below, including Japanese-language reporting where cited, then editorially reviewed before publishing.