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

In collaboration with the Metal engineering team at Apple, PyTorch today announced that its open source machine learning framework will soon support GPU-accelerated model training on Apple silicon Macs powered by M1, M1 Pro, M1 Max, or M1 Ultra chips.

pytorch
Until now, PyTorch training on the Mac only leveraged the CPU, but an upcoming version will allow developers and researchers to take advantage of the integrated GPU in Apple silicon chips for "significantly faster" model training.

A preview build of PyTorch version 1.12 with GPU-accelerated training is available for Apple silicon Macs running macOS 12.3 or later with a native version of Python. Performance comparisons and additional details are available on PyTorch's website.

Tag: Metal

Popular Stories

iPhone SE 4 Vertical Camera Feature

iPhone SE 4 Production Will Reportedly Begin Ramping Up in October

Tuesday July 23, 2024 2:00 pm PDT by
Following nearly two years of rumors about a fourth-generation iPhone SE, The Information today reported that Apple suppliers are finally planning to begin ramping up mass production of the device in October of this year. If accurate, that timeframe would mean that the next iPhone SE would not be announced alongside the iPhone 16 series in September, as expected. Instead, the report...
iPhone 17 Plus Feature

iPhone 17 Lineup Specs Detail Display Upgrade and New High-End Model

Monday July 22, 2024 4:33 am PDT by
Key details about the overall specifications of the iPhone 17 lineup have been shared by the leaker known as "Ice Universe," clarifying several important aspects of next year's devices. Reports in recent months have converged in agreement that Apple will discontinue the "Plus" iPhone model in 2025 while introducing an all-new iPhone 17 "Slim" model as an even more high-end option sitting...
Generic iPhone 17 Feature With Full Width Dynamic Island

Kuo: Ultra-Thin iPhone 17 to Feature A19 Chip, Single Rear Camera, Semi-Titanium Frame, and More

Wednesday July 24, 2024 9:06 am PDT by
Apple supply chain analyst Ming-Chi Kuo today shared alleged specifications for a new ultra-thin iPhone 17 model rumored to launch next year. Kuo expects the device to be equipped with a 6.6-inch display with a current-size Dynamic Island, a standard A19 chip rather than an A19 Pro chip, a single rear camera, and an Apple-designed 5G chip. He also expects the device to have a...
iPhone 16 Pro Sizes Feature

iPhone 16 Series Is Less Than Two Months Away: Everything We Know

Thursday July 25, 2024 5:43 am PDT by
Apple typically releases its new iPhone series around mid-September, which means we are about two months out from the launch of the iPhone 16. Like the iPhone 15 series, this year's lineup is expected to stick with four models – iPhone 16, iPhone 16 Plus, iPhone 16 Pro, and iPhone 16 Pro Max – although there are plenty of design differences and new features to take into account. To bring ...
icloud private relay outage

iCloud Private Relay Experiencing Outage

Thursday July 25, 2024 3:18 pm PDT by
Apple’s iCloud Private Relay service is down for some users, according to Apple’s System Status page. Apple says that the iCloud Private Relay service may be slow or unavailable. The outage started at 2:34 p.m. Eastern Time, but it does not appear to be affecting all iCloud users. Some impacted users are unable to browse the web without turning iCloud Private Relay off, while others are...
iPhone 17 Plus Feature Purple

iPhone 17 Rumored to Feature Mechanical Aperture

Tuesday July 23, 2024 9:32 am PDT by
Apple is planning to release at least one iPhone 17 model next year with mechanical aperture, according to a report published today by The Information. The mechanical system would allow users to adjust the size of the iPhone 17's aperture, which refers to the opening of the camera lens through which light enters. All existing iPhone camera lenses have fixed apertures, but some Android...

Top Rated Comments

innominato5090 Avatar
29 months ago
So unbelievably excited about this. Being able to test GPU-accelerated code locally (and even train some smaller networks) rather than having to rely on a unix-server will speed up ML development significantly.

Can't wait for it to become mature enough to be merged in a stable branch.
Score: 13 Votes (Like | Disagree)
mukiex Avatar
29 months ago
Personally I'm hoping to see support for the neural engine in the future. When they added NE support to Topaz, my M1 Macbook Air suddenly started performing on par with my desktop GTX 1080.
Score: 5 Votes (Like | Disagree)
mrWONERFUL Avatar
29 months ago
I hope to be blown away with what wwdc has to announce
Score: 4 Votes (Like | Disagree)
adamw Avatar
29 months ago
Always good to see move native support for Metal in MacOS on M1 family of chips...
Score: 3 Votes (Like | Disagree)
seek3r Avatar
29 months ago
Nice step in the direction of finding out how capable these GPUs are for this kind of workload, may also give an idea if they work well enough for GPGPU and ML work to infer if Apple is likely to develop their own GPU for the eventual AS Mac Pro
Score: 3 Votes (Like | Disagree)
Luca1995it Avatar
29 months ago
I started collecting benchmarks of M1 Max on PyTorch here: https://github.com/lucadiliello/pytorch-apple-silicon-benchmarks
Score: 2 Votes (Like | Disagree)