Benchmarked: Apple Core AI vs MLX for on-device LLM speed on iPhone and Mac

HankYeomans · x · 2026-09-08

A new hands-on benchmark compares Apple's Core AI framework against the open-source MLX stack, running the same model on both iPhone and Mac to see which delivers faster on-device inference. The piece focuses on real speed differences across the two devices and frameworks.

Original post →

More from Infra

Infra channel →