Search
NVIDIA shifts to autonomous server production, with OEM manufacturers taking a secondary role
1 week ago /
Read about 0 minute
Author：小编
NVIDIA plans to revolutionize the way high-performance AI servers are built through its 'Vera Rubin' product line. This product line will enter mass production in the form of the VR200 series by the end of 2026. NVIDIA will no longer solely supply CPUs and GPUs to OEM manufacturers. Instead, it will provide complete L10 compute trays pre-installed with 'Vera' CPUs, 'Rubin' accelerators, memory, 800G network cards, 110kW power supplies, and liquid cooling infrastructure, with all components verified before leaving the factory.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
