On August 25, 2026, Apple announced two new chips: M6 and M5 Ultra. M6 is the company’s first chip built on a 2nm process, debuting in the new Mac mini. M5 Ultra is Apple’s first quad-die design and its most powerful chip ever, shipping in the new Mac Studio. Pre-orders opened the same day; general availability starts September 22, 2026 across 30 countries and regions. The shared message is easy to read: on-device AI is the next hardware battleground.
M6: The First 2nm
M6 pairs a 12-core CPU (2 super, 4 performance, 6 efficiency cores — two more than M5) with a 12-core GPU that puts a Neural Accelerator in every core. Apple claims the world’s fastest single-threaded performance, with multithreaded gains of up to 1.2x over M5 and 2.4x over M1. The Neural Engine is a dual 16-core design delivering up to 2x peak AI compute versus previous generations, and peak GPU AI compute is nearly 30% higher than M5.
The memory story targets mainstream developers: unified memory tops out at 32GB with 170GB/s of bandwidth — 10% more than M5 and 2.5x the M1. For anyone running local models on a laptop or a small chassis, that bandwidth figure matters more than the core count, because token generation for local LLMs is largely a memory-bandwidth problem.
M5 Ultra: Four Dies, 512GB of Memory
M5 Ultra is the bigger story for AI work. Apple fuses two dual-die M5 Max chips into a single quad-die package through the next generation of UltraFusion, with inter-die bandwidth above 4.4TB/s. The CPU scales to 36 cores (12 super, 24 performance), up to 1.25x faster single-threaded and 1.3x faster multithreaded than M3 Ultra. The GPU reaches 80 cores, again with per-core Neural Accelerators, for up to 4.5x the peak GPU AI compute of M3 Ultra and up to 40% faster graphics.
The number that matters most is memory: up to 512GB of unified memory with 1.2TB/s of bandwidth, 50% more than M3 Ultra. Apple says outright that this is enough to run LLMs with hundreds of billions of parameters on the desk. That is a deliberate shot at the shared-memory ceiling — model weights have to fit in memory to run locally, and 512GB covers model classes that previously implied a multi-GPU server rack. The Mac Studio lineup starts at $2,499 with M5 Max (18-core CPU, up to 40-core GPU, up to 128GB memory, 614GB/s bandwidth) and $5,499 with M5 Ultra; the 512GB configuration arrives in late October.
One Desk, One Small Cluster
The sleeper feature for AI engineers is clustering: over Thunderbolt 5 with RDMA, four Mac Studio units can be linked into a single system for up to 3x faster distributed AI inference. Add 120Gb/s Thunderbolt 5, first-time Wi-Fi 7 and Bluetooth 6 via Apple’s N1 chip, support for up to eight displays, and a media engine with hardware AV1 decode and up to 33 simultaneous 8K ProRes 422 streams, and the target user is written on the spec sheet. This is a desktop designed around local model workflows — prompt processing for large LLMs, agentic workloads, and local fine-tuning through Core AI, Core ML, Metal, and Xcode.
The Reality Check for On-Device AI
Apple’s pitch is concrete: faster LLM prompt processing, agentic workloads, and the ability to run — and fine-tune — large models locally, alongside macOS 27 “Golden Gate” and expanded Apple Intelligence features. The strategy positions privacy, a one-time purchase, and no token bills as the alternative to cloud APIs.
The reality check is equally concrete. The 512GB configuration ships in late October, and $5,499 is not an entry-level price. The “up to 4.3x AI performance” claim is measured against the previous generation, and turning peak numbers into real local inference quality depends on framework and model optimization that takes time. Quantized open models already run well on this class of hardware; the open question is how much of the cloud-API workflow — tooling, serving, fine-tuning — matures on the desktop too. But the direction is unambiguous: when a flagship desktop can hold a hundred-billion-parameter model in memory, running large models on-device moves from demo to purchase order — and that is a longer-term pressure on the cloud inference market than any keynote slogan.
Sources
- Apple introduces M6 and M5 Ultra — Apple Newsroom
- Apple introduces new Mac Studio with M5 Max and M5 Ultra — Apple Newsroom
AI-assisted summary compiled from the sources above, reviewed by a human before publishing.