Apple May Return to Server Market with Nvidia Technology

Apple is weighing a return to the server business with a dedicated AI system built around its M8-series chips, intended for sale to external customers and aimed at inference and response generation from trained models. Discussions have included using Nvidia’s NVLink Fusion to provide high-speed chip-to-chip networking and software for linking multiple M8 processors, and the initiative reportedly has backing from John Ternus. A 2029 launch is being targeted, though the effort remains unfinished. The move follows strong enterprise demand for Mac mini and Mac Studio hardware and stems in part from limitations in scaling Apple’s existing Private Cloud Compute infrastructure, which the company currently reserves for internal Apple Intelligence tasks and has declined to open to partners.

Bringing Nvidia networking into an Apple-branded server would give Nvidia a role in a product that could compete with its own AI systems, expanding NVLink Fusion’s customer base beyond Nvidia-chip deployments. The talks reflect a thaw in Apple-Nvidia relations; Apple is already extending Private Cloud Compute to Google Cloud using Nvidia GPUs and has explored other collaborative options, including Nvidia’s open-source models. Significant obstacles remain, notably the need for stronger enterprise support, more software tooling for AI developers, and further investment in Apple’s MLX framework, leaving the project contingent on resolving these software and business gaps.