Accelerate your machine learning workloads with the M5 and A19 GPUs | Apple Developer @AppleDeveloper
Accelerate your machine learning workloads with the M5 and A19 GPUs | Apple Developer @AppleDeveloper
Uploaded March 2026 | Updated September 2026, 1 week ago
Discover how to take advantage of the M5 and A19 GPUs to accelerate machine learning. Find out how to use the Neural Accelerators inside each GPU core to perform accelerated tensor and matrix operations for AI training and inference, and learn how to optimize performance using the Metal debugger and Metal system trace.

Download the Metal Performance Primitives (MPP) Programming Guide: apple.co/MPPGuide_Download

More Apple Developer resources:
Video sessions: apple.co/VideoSessions
Documentation: apple.co/DeveloperDocs
Forums: apple.co/DeveloperForums
App: apple.co/DeveloperApp
Accelerate your machine learning workloads with the M5 and A19 GPUs | Apple DeveloperWWDC26: Run local agentic AI on the Mac using MLX | AppleWWDC26: Communicate your brand identity on iOS | AppleWWDC26: Use foveated streaming to bring immersive content to visionOS | AppleWWDC26: What’s new in Apple In-App Purchase | AppleWWDC26: Make your game great with touch | AppleWWDC26: Xcode Tips and Tricks Group Lab | AppleWWDC26: Rev up your CarPlay app | AppleWWDC26: Platforms State of the Union | AppleSwiftUI essentials | Meet with AppleDub Dub Daily: Day 4 | WWDC26WWDC26: Collaborate on structured 3D models in visionOS | Apple
Apple Developer |

Accelerate your machine learning workloads with the M5 and A19 GPUs | Apple Developer

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER