Superpower Daily
Everlaw Connects Gemini Legal to Governed Evidence as Weil Depl...Stanford Finds a 19% Employment Gap for Young Workers in AI-Exp...Nvidia’s NeMo Switchyard Routes Agent Calls Across Models, Not...Guidelight Finds Few Public Rogue-AI Plans as California Requir...HiDream-O1-World Tops WBench Navi at 80.9, Betting on Persisten...Apple Refreshes Mac mini and Mac Studio for Linked Local AI, Fr...Everlaw Connects Gemini Legal to Governed Evidence as Weil Depl...Stanford Finds a 19% Employment Gap for Young Workers in AI-Exp...Nvidia’s NeMo Switchyard Routes Agent Calls Across Models, Not...Guidelight Finds Few Public Rogue-AI Plans as California Requir...HiDream-O1-World Tops WBench Navi at 80.9, Betting on Persisten...Apple Refreshes Mac mini and Mac Studio for Linked Local AI, Fr...
Productspublished

Apple Refreshes Mac mini and Mac Studio for Linked Local AI, From $899

The specification refresh formalizes a workflow of linking Macs for larger local models, but performance evidence has yet to arrive and the top chip’s name is inconsistent.

By 2 min read
Apple Refreshes Mac mini and Mac Studio for Linked Local AI, From $899
Sign in to save and react

Listen to this story

The audio brief

About 1:43
0:001:43
Read transcript
Apple is making memory capacity and machine-to-machine links the main event in its refreshed Mac mini and Mac Studio, putting local AI inference ahead of new consumer features. The entry point is an $899 Mac mini with Apple’s M6 chip, a 12-core CPU, a 12-core GPU, and a maximum of 32 gigabytes of unified memory. Apple says its multithreaded CPU performance is up to 40 percent faster than the M4, but independent benchmarks were not available at launch. The more ambitious idea is to link several Macs together. Apple’s MLX framework uses the unified memory in Apple silicon to run models locally, while macOS 26.2 adds low-latency communication between Thunderbolt 5 hosts. In practical terms, developers, researchers, and hobbyists can pool multiple Mac minis or Studios to run language models too large for one mass-market machine. At the top, the $5,499 Mac Studio is described with up to 512 gigabytes of unified memory and memory bandwidth reaching 1.2 terabytes per second. Both desktops add Wi-Fi 7 and Bluetooth 6 through Apple’s N1 networking chip. The mini includes 2.5-gigabit Ethernet, with 10-gigabit available as an option. Preorders opened August 25, shipments start September 22, and the 512-gigabyte configuration arrives in late October. One basic detail remains unresolved: launch materials call the flagship chip M5 Ultra, while another hardware description calls it M6 Ultra. Until testing arrives, capacity is the clearest selling point—and the performance case is still unproven.

Story brief

3 key points

Apple is using refreshed Mac mini and Mac Studio hardware to make memory capacity and machine-to-machine links the main local-AI selling points. The $899 M6 Mac mini tops out at 32GB, while the $5,499 Studio is described with up to 512GB of unified memory and 1.2TB/s bandwidth—more relevant to running larger models than adding consumer features. MLX and Thunderbolt 5 networking enable multi-Mac inference, but...

  1. 01

    macOS 26.2 adds low-latency Thunderbolt 5 host communication for distributed MLX inference across linked Macs.

  2. 02

    Apple claims M6 Mac mini CPU performance up to 40% over M4; independent benchmarks were unavailable at launch.

  3. 03

    Both desktops support Wi-Fi 7 and Bluetooth 6; Mac mini includes 2.5Gb Ethernet, with 10Gb optional.

Apple’s new Mac mini and Mac Studio put local AI inference at the center of a desktop refresh that otherwise adds no major features. The bet is on memory, bandwidth and linked-machine workloads rather than a new AI application for the Mac.

Apple is positioning the desktops for local and distributed inference using MLX, an open-source array framework built to use Apple silicon’s unified memory. The approach gives developers a way to run models on their own hardware rather than depend entirely on cloud services.

Thunderbolt links turn desktops into a larger pool

macOS 26.2 added low-latency communication between Thunderbolt 5 hosts for distributed MLX inference. Developers, researchers and hobbyists have used that capability to chain Mac minis or Studios together, creating enough combined capacity for local language models that would not fit on a single mass-market machine.

The entry point and the memory ceiling
$899M6 Mac mini

The M6 Mac mini starts at $899.

$5,499M5 Ultra Mac Studio

The Mac Studio with M5 Ultra starts at $5,499.

512GBTop Studio capacity

The highest-end Studio configuration is described as supporting up to 512GB of unified memory.

Capacity is the product feature

At the lower end, the M6 has a 12-core CPU, 12-core GPU, up to 160GB per second of unified-memory bandwidth and a 32GB memory maximum. Apple says its multi-threaded CPU performance is up to 40% faster than M4, though independent benchmarks were not yet available.

  • The top-end Studio is described as reaching 512GB of unified memory and up to 1.2TB per second of memory bandwidth, an Apple figure aimed especially at AI workloads.
  • Both desktops include Apple’s N1 networking chip with Wi-Fi 7 and Bluetooth 6 support. Apple says storage can reach 15GB per second.
  • The Mac mini includes 2.5Gb Ethernet as standard, with a 10Gb option.

One detail needs clarification. The launch and pricing details call the high-end chip M5 Ultra, while the later hardware description calls it M6 Ultra. The capacity and bandwidth figures are clear, but the conflicting designation leaves the flagship chip’s name unresolved.

Orders are open before testing begins

Preorders began August 25, and Apple plans to ship both desktops September 22. A 512GB memory configuration is scheduled for late October, leaving early buyers to weigh the published specifications before independent performance testing is available.

Sources

  1. arstechnica.comWith new Mac Studio and Mac mini, Apple leans hard into local AI inference

Related coverage

Continue reading