Apple reportedly developing M8 Ultra-based server for AI inference
Apple is reportedly working on an enterprise server with two or four M8 Ultra chips designed for AI inference, targeting developers, businesses, and governments. The company is considering Nvidia's NVLink Fusion technology to connect the chips, with a possible launch no earlier than 2029. The project may gain momentum as OpenAI and Anthropic already purchase Mac hardware in bulk for AI workloads.
Apple’s reported server would pair its custom M8 Ultra silicon in two- or four-chip configurations, explicitly optimized for inference rather than training. The company is evaluating Nvidia’s NVLink Fusion interconnect to link the chips, a technology originally proprietary but now open to third parties. The timeline is tentative—no earlier than 2029—and the project remains subject to cancellation or a shift away from Nvidia’s solution. CEO John Ternus reportedly championed the effort during his hardware tenure, seeing a strategic opening in the surging AI compute market.
The initiative gains context from recent purchasing patterns: OpenAI and Anthropic have been acquiring Mac Minis and Mac Studios in bulk for AI workloads, and Apple’s Mac revenue climbed nearly 29% last quarter to $10.4 billion. This suggests existing demand for Apple silicon in AI settings, even if the enterprise server is a more ambitious, data-center-grade step. The report, sourced from The Information, indicates Apple is positioning itself as a serious contender in AI infrastructure, though the long lead time and technical hurdles remain significant.
If realized, Apple’s server could diversify the AI inference hardware market, currently dominated by Nvidia’s GPUs. Developers, businesses, and governments might gain an alternative that leverages Apple’s energy-efficient chip design, potentially lowering operational costs for running trained models. However, the 2029 timeline and reliance on Nvidia’s interconnect mean adoption is uncertain. The project could also pressure Nvidia to innovate further, or it may simply validate existing Mac-based AI workflows. Ultimately, the impact hinges on execution, pricing, and whether Apple commits fully to enterprise infrastructure.