Apple Develops M8 Ultra AI Server With NVIDIA NVLink Fusion
Apple designing servers around NVIDIA's interconnect points to an enterprise hardware push rather than a closed in-house stack, and it puts the company's AI infrastructure work on a reported commercial timeline.
Reporting from 1 source: GIGAZINE.
Apple is developing an AI server built around M8 Ultra chips connected through NVIDIA's NVLink Fusion, according to The Information. Two configurations are in development, one with two M8 Ultras and one with four. Apple plans to sell the servers to enterprise customers, with launch reported around 2029. Manufacturing would involve TSMC's N3P process, Broadcom, and Foxconn.
The reported server design pairs Apple silicon with NVIDIA's NVLink Fusion, a platform for connecting outside CPUs and custom accelerators to NVIDIA's NVLink. That connection would let Apple silicon plug into NVIDIA's rack-scale AI systems.
Chip development is split across partners. TSMC's N3P process is named for manufacturing, Broadcom supports Apple's chiplet designs, and Foxconn handles server assembly, with design support from Lenovo and its subsidiary. Apple would sell the finished servers to enterprise customers.
- Two M8 Ultra configuration: One of two servers in development, equipped with two M8 Ultra chips
- Four M8 Ultra configuration: The second server in development, equipped with four M8 Ultra chips
Synthesized by Yomimono from the 1 cited source below, including Japanese-language reporting where cited, then editorially reviewed before publishing.