Machine Learning Framework PyTorch Enabling GPU-Accelerated Training on Apple Silicon Macs

[MacLC]

Are you referring to the M1 GPU vs CPU improvements or something else? The competition is not really the CPU version, but other dedicated GPUs, likely Nvidia.

There are M1 Ultra benchmarks on that site.

Time will tell. I doubt the M2 Ultra will be a huge jump, but ultimately Apple will need a real Mac Pro replacement, so that will be the one to aim for. I’ll keep an eye on things, maybe train a few smaller models on the Apple side, but so far it’s not even close to replace any of my Nvidia systems, not even the smallest one.

This is a slightly late reply...

The link I didn't post was:

Running PyTorch on the M1 GPU

Today, the PyTorch Team has finally announced M1 GPU support, and I was excited to try it. Here is what I found.

sebastianraschka.com

I fully agree with your assessment, though. Especially with the Ethereum merge, high end Nvidia cards have gotten very, very cheap. Even the 2080 Sebastian tested was by itself 2x the speed of the 48-core M1 Ultra.

As an update, I've got a machine on the side that I reserve for Premiere but got a crypto crash 3080 Ti for it...

You're totally right. It's night and day faster than Apple Silicon. It's actually embarrassingly faster.