Apple is planning a major return to the enterprise server market, a segment it abandoned nearly two decades ago. This move targets AI developers and government clients with a new machine designed specifically for inference workloads. The shift signals a strategic pivot toward high-performance computing infrastructure after years of focusing on consumer and professional desktop hardware.
Apple targets enterprise AI inference with M8 Ultra chips and potential Nvidia interconnect
The core of this new system is the M8 Ultra chip, which Apple describes as its most powerful silicon to date. The company is currently in the research and development phase for this processor. It will serve as the foundation for a new line of AI servers aimed at heavy computational tasks.
Apple intends to offer two distinct hardware configurations for this server. One version will house two M8 Ultra chips, while another will support four of them. To manage data flow between these processors, the company is exploring Nvidia's NVLink Fusion technology. This interconnect solution would allow the chips to communicate at high speeds, a critical requirement for large-scale AI operations.
The project remains subject to change and could be cancelled before reaching production. The server is not expected to launch before 2029. This timeline reflects the long development cycle for custom silicon and the complexity of integrating third-party interconnect technologies. We looked at previous Apple server attempts while tracking this development.
Discussion
0 comments
Log in to join the thread with a thoughtful take, question, or correction.