How Fast Is MLX? A Comprehensive Benchmark on 10 Apple Silicon Chips and 3 CUDA GPUs

A benchmark of the main operations and layers on MLX, PyTorch MPS and CUDA GPUs.

Image by author: Example of benchmark on the softmax operation

In less than two months since its first release, Apple’s ML research team’s latest creation, MLX, has already made significant strides in the ML community. It is remarkable to see how…