Apple M6

Starting from local compute, see how Apple M6 and M5 Ultra rewrite on-device large-model work through unified memory and bandwidth.

Apple updated M6 and M5 Ultra at once, framing local AI around memory capacity and bandwidth rather than CPU scores. M6 brings 2nm to the M series with dual 16-core Neural Engines, up to 170GB/s of bandwidth and 32GB of unified memory; M5 Ultra links four dies for 1.2TB/s and up to 512GB, letting Mac Studio run hundred-billion-parameter models on device. When memory runs short, models shrink or fall back to the cloud.

New Mac mini and Mac Studio on sale today: M6 delivers 4x on-device AI

On September 22, 2026, Apple announced via its online newsroom that the new Mac mini and Mac Studio are officially on sale today. Following their Augu...

Admin 9/22/2026

Apple released the M6 and M5 Ultra, and Mac's overall push into local AI computing power

Apple released two chips, the M6 and M5 Ultra , which entered the new Mac mini and Mac Studio respectively. The former introduced the 2nm process into...

Admin 8/26/2026

113

Apple M6 and M5 Ultra released: Local AI computing power begins to compete for unified memory

On August 25, 2026, Apple announced the M6 and M5 Ultra in a press release, focusing not just on switching chips to a new generation but on bringing l...

Admin 8/25/2026

121