Apple Debuts 2nm M6 and Quad-Die M5 Ultra Chips for Local AI Compute

Apple debuted the M6 and M5 Ultra chips on August 25, 2026, integrating them into new Mac mini and Mac Studio models. [2] The M6 is Apple's first processor built on a 2-nanometer process and supports up to 32GB of unified memory. [2] The Mac mini starts at $899, representing a $100 price increase over its predecessor, and features either the new M6 or M5 Pro processor. [3] This product launch is likely the final one under the tenure of CEO Tim Cook. [3]

The M5 Ultra chip introduces Apple's first quad-die design, utilizing UltraFusion technology to connect two dual-die M5 Max chips. [2] It supports up to 512GB of unified memory, allowing the Mac Studio to run large language models with hundreds of billions of parameters entirely on device. [2] Both the Mac mini and Mac Studio have become popular among consumers running autonomous AI agents locally, such as OpenClaw. [3]

Apple's move to a quad-die M5 Ultra and a 512GB unified memory ceiling directly targets the high-end, local AI compute market. By offering machines capable of running massive language models on-device, Apple positions the Mac Studio as a workstation for practitioners running autonomous agents like OpenClaw. The M5 Ultra boasts 50 percent more memory bandwidth than its predecessor, the M3 Ultra, marking a significant leap in data throughput. Furthermore, the release highlights a major transitional moment for Apple as Tim Cook's tenure likely nears its end. What the sources don't address: How the thermal performance of the new quad-die M5 Ultra will be managed under sustained AI agent workloads.