Minions
A dynamic, distributed virtual GPU cloud that aggregates idle Apple Silicon GPUs across macOS devices to run large-scale machine-learning inference workloads.
Description
Minions is a dynamic, distributed virtual GPU cloud developed by Apple Machine Learning Research. It aggregates idle Apple Silicon GPUs from macOS devices across an organization and exposes them as a shared resource pool for executing machine-learning inference workloads.
The goal is to work around the limitations of centralized GPU infrastructure — high cost, limited elasticity and long wait times under resource constraints — by leveraging hardware teams already own, without procuring additional accelerators.
The system provides flexible GPU inference task execution, fault tolerance and efficient resource utilization. Minions is described in an Apple research paper published in October 2026; its authors are Jianhua Li, Yun Xing, Girisha De Silva, Ryan Flint, Kai Yuan and Sean Suchter.