mobile_app_open icon indicating copy to clipboard operation
mobile_app_open copied to clipboard
verified publisher icon mlcommons

iPhone TOPS estimation

Open freedomtan opened this issue 2 months ago • 1 comments

I developed a simple program that utilizes Apple’s Metal Performance Shaders Graph (MPSGraph) to measure and estimate the compute capacities of Apple Neural Engine (ANE). The program can be found on GitHub at https://github.com/freedomtan/measure_ane_capacity. By using this program, I obtained numbers that are quite close to what Apple claimed.

Based on my observations, I believe that to achieve better performance, we may need to convert TensorFlow (TF) or PyTorch models to MPSGraph (instead of CoreML).

freedomtan avatar Dec 31 '25 03:12 freedomtan

Why we cannot get these numbers with Core ML tools? It seems, for either the palette or linear quantization schemes, the models are actually using float16 as we knew before.

freedomtan avatar Jan 06 '26 02:01 freedomtan