Geekbench Machine-Learning Benchmark Tools Add Support for macOS and Other Desktop Platforms

[MacRumors]

Geekbench developer Primate Labs has announced a new 0.6 preview version of Geekbench ML that expands beyond iOS and Android to offer benchmarking capabilities for AI and other machine learning workflows on desktop platforms, including macOS, Windows, and Linux.

AI and ML-related workflows aren’t just confined to mobile, and hardware architecture on desktop and laptop devices is changing to accommodate this shift in computing. With this latest 0.6 preview, Geekbench ML now supports Windows, macOS, and Linux. This means you’ll be able to see how machine learning-powered tasks run on your desktop, laptop, or even a server — whether it has new AI-specific hardware or not. And, as always, our models and data sets are identical across all supported platforms, making scores comparable.

With new and improved frameworks, Geekbench ML supports faster performance on the latest hardware using more up-to-date models including for Core ML on iOS and macOS.

New workloads for Geekbench ML include Depth Estimation to reflect capabilities related to software-assisted portrait mode photo effects, Style Transfer to measure the ability to transform one image into the style of another one, and Image Super-Resolution to assess AI-driven image upscaling.

Geekbench ML 0.6 is still a preview release that Primate Labs will continue to tweak based on results and feedback, so scores from version 0.6 can't be directly compared to those from the previous 0.5 release. Primate Labs is looking toward a polished 1.0 release at some point in 2024.

Geekbench ML is available in the App Store and Google Play Store, and Primate Labs is directly hosting preview versions for macOS, Windows, and Linux on its website.

Article Link: Geekbench Machine-Learning Benchmark Tools Add Support for macOS and Other Desktop Platforms

 

[Adreitz]

Nice. Hopefully this will allow more direct comparisons between M-series processors and PC GPUs for ML tasks. I enjoy using SDXL on my M2 Max, but it's not very fast.

 

[Blue Sun]

I’m interested to see how much better the M3 Max chip is compared to M1 and M2. Apple claimed big gains in ML tasks using the new chips.

 

[Chuckeee]

I wonder whose chipset this benchmark is optimized for.

 

[high3r]

My M1 Pro MBP's neural engine scores higher than A17 Pro. According to Apple the former is 11 TOPS while the latter is 35 TOPS. Either Apple is lying or this benchmark is worthless atm.