MLX does not run on NPUs AFAIK; just gpu and cpu. You have to use CoreML to offi... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		nullstyle 32 days ago \| parent \| context \| favorite \| on: Inside the M4 Apple Neural Engine, Part 1: Reverse... MLX does not run on NPUs AFAIK; just gpu and cpu. You have to use CoreML to officially run code on the neural engine.

mirsadm 32 days ago [–]

Even then there is no transparency on how it decides what runs on the ANE/GPU etc

sroussey 32 days ago | [–]

Correct. OS level stuff get first priority, so you can’t count on using it.

znagengast 32 days ago | | [–]

Turns out third party actually gets priority for ANE

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact