OFICIAL Apple Newsroom

Apple introduces M6 and M5 Ultra for a big leap in performance and AI compute

What happened
Based on Apple Newsroom · Aug 25, 2026

Apple introduced the M6 and M5 Ultra chips, featuring advanced 2 nm and quad-die architectures, respectively, to enhance desktop performance and AI capabilities in new Mac mini and Mac Studio models.

🚩 Report a correction ↗ Open source
Apple introduces M6 and M5 Ultra for a big leap in performance and AI compute
Apple Newsroom — Apple
Key points
·
M6, Apple’s first state-of-the-art 2-nanometer chip, advances every compute block, delivering gains across every dimension of performance.
·
The chip features a larger 12-core CPU complex with the world’s fastest CPU core, a larger 12-core GPU with Neural Accelerators, a Dual 16-core Neural Engine, and up to 170GB/s of unified memory bandwidth.
·
M5 Ultra, the ultimate powerhouse for pro and AI workloads, uses next-generation UltraFusion technology to form a quad-die architecture for the first time in an M-series system on a chip (SoC).
·
The chip includes an up-to-36-core CPU and up-to-80-core GPU with a massive 1.2TB/s of unified memory bandwidth, 50 percent more than M3 Ultra.
Key numbers
·
2x faster multithreaded performance than M5 and supports up to 32GB of unified memory with 170GB/s bandwidth.
·
The M5 Ultra, introduced in the new Mac Studio, uses a quad-die architecture with UltraFusion technology, connecting two dual-die M5 Max chips for the first time.
·
2TB/s of unified memory bandwidth and support for up to 512GB of memory.

Apple unveiled the M6, its first 2 nm chip, in the new Mac mini, featuring a 12-core CPU, 12-core GPU, and a Dual 16-core Neural Engine. The chip delivers up to 1.2x faster multithreaded performance than M5 and supports up to 32GB of unified memory with 170GB/s bandwidth. It is designed for everyday tasks, coding, and on-device AI workflows, offering improved energy efficiency and faster model execution.

The M5 Ultra, introduced in the new Mac Studio, uses a quad-die architecture with UltraFusion technology, connecting two dual-die M5 Max chips for the first time. It includes up to a 36-core CPU and 80-core GPU, with 1.2TB/s of unified memory bandwidth and support for up to 512GB of memory. The chip targets professionals handling complex workloads like 3D rendering, AI models, and high-resolution video editing.

Apple highlighted the M6’s Dual 16-core Neural Engine, which provides up to 2x peak compute over previous generations, enabling faster on-device AI tasks. The GPU includes Neural Accelerators, delivering nearly 30% more peak GPU compute for AI compared to M5. The chip also supports hardware-accelerated ray tracing and Dynamic Caching for improved graphics performance.

Apple emphasized the M5 Ultra’s capabilities for AI and graphics-intensive workloads, including a 32-core Neural Engine and next-generation GPU with Neural Accelerators. The chip’s 1.2TB/s memory bandwidth and 512GB unified memory support allow running large AI models locally. Developer tools like Core AI, Core ML, and Metal are optimized to leverage these advancements for on-device AI and graphics processing.

Original source → Deals on Clipraptor.com →