Apple Unveils 2nm M6 and Quad-Die M5 Ultra

On August 25, 2026, Apple announced its next-generation chipsets, the M6 and M5 Ultra, declaring a leap in desktop performance and AI computing capabilities. The M6 is Apple's first 2nm process chip, optimized for a wide range of users from everyday tasks to developers and AI hobbyists.

The M6 features a 12-core CPU (2 super cores, 4 performance cores, 6 efficiency cores), a 12-core GPU, and a dual 16-core Neural Engine. Peak compute performance has improved by up to 2x compared to the previous generation, and integrated memory bandwidth has increased to up to 170GB/s. Single-thread performance has reached world-class levels, while multi-thread performance is up to 1.2x faster than the M5 and up to 2.4x faster than the M1.

The M5 Ultra is the most powerful chip for professional and AI workloads, being the first M-series SoC in Apple's history to adopt a Quad-die architecture. Utilizing next-generation UltraFusion technology, it configures up to a 36-core CPU and 80-core GPU, with integrated memory bandwidth reaching 1.2TB/s. This is 50% higher bandwidth than the M3 Ultra, optimized for running large AI models and complex data processing.

These chipsets are featured in the new Mac mini and Mac Studio, respectively, maximizing performance while maintaining energy efficiency in various professional applications such as Adobe Photoshop, Xcode, Adobe Premiere Pro, and LM Studio.