Apple's New Chips: The M5 Ultra and M6, Explained Simply

Apple announced two new computer chips on August 25, 2026: the M5 Ultra and the M6. These chips power updated versions of the Mac Mini and Mac Studio, Apple's small desktop computers. Apple called the M5 Ultra its "most powerful chip ever" TechCrunch.

Think of a computer chip as the brain of a device. The M5 Ultra is Apple's first chip to combine four separate brain pieces into one package. Until now, Apple's most powerful chips combined two. This one combines four, which gives it a 36-core CPU (the part that handles general thinking), an 80-core GPU (the part that handles graphics and visual tasks), and a very large amount of fast memory — 1.2 terabytes per second of bandwidth, which is 50% more than the previous Ultra chip TechCrunch. That matters for artificial intelligence, because AI programs need to hold huge amounts of information in fast memory to work well.

The M6 is a different kind of chip. It is built using a 2-nanometer process, which means the tiny components on the chip are packed more tightly than before, making it faster and more efficient. It has 12 CPU cores (two more than the M5) and a Dual 16-core Neural Engine, which is a specialized part of the chip designed specifically for AI tasks. Apple says the M6 delivers nearly 30% better AI performance than the M5, especially for running AI text-generation programs directly on the computer TechCrunch. Sri Santhanam, Apple's vice president of Silicon Engineering, commented on the M6 in the company's press release TechCrunch; Apple Newsroom.

The new Mac Studio, which comes with either the M5 Max or M5 Ultra chip, is up to 4.3 times faster at AI tasks than the previous model Stocktitan. The Mac Mini now comes in versions powered by either the M6 or the M5 Pro, depending on the model. Pre-orders for the Mac Mini start at $899, and units ship after September 22, 2026 TechCrunch.

These launches complete a very busy year for Apple's chip development, even by Apple's standards. The base M5 chip arrived in October 2025 as a leap in AI performance Apple Newsroom. The 14-inch MacBook Pro with M5 followed, with up to 1.6x faster graphics in professional applications Apple Newsroom. In March 2026, Apple released the M5 Pro and M5 Max for the MacBook Pro, with an up-to-18-core CPU featuring six "super cores" that Apple calls the world's fastest, alongside 12 performance cores Apple Newsroom. Those chips pair an advanced CPU with a next-generation GPU containing Neural Accelerators and higher unified memory Apple Newsroom. The MacBook Air received the base M5 that same month, delivering up to 6.9x faster AI video enhancement Apple Newsroom. The 14-inch MacBook Pro with M5 now ships standard with 1TB of storage in space black and silver, starting at $1,699 Apple Newsroom.

The M5 Ultra's four-in-one design is the biggest architectural change. By combining four chip pieces instead of two, Apple can offer a large enough memory pool to run big AI programs that would otherwise need separate, expensive graphics cards.

The M6's 2-nanometer manufacturing process is the other key detail. Apple has not said which company actually manufactures the chip, but the tighter component packing, combined with the expanded Neural Engine and extra GPU cores, points to where Apple sees things heading: AI programs running directly on your computer, responding faster, rather than relying on the internet.

What is less clear is how the M5 Ultra and M6 compare to each other for AI tasks. Apple describes the Ultra as the heavy-duty option for the most demanding work and the M6 as the efficient everyday choice, but the company has not published head-to-head tests. Developers will need to wait for independent reviews to know which chip is better for a given AI program.

The pricing is notable. A Mac Mini starting at $899 with the M6 chip puts very current, AI-focused technology into a price range that many more people can afford. For developers building AI applications that run on a user's own computer, that is a low barrier to getting started.

In my view, Apple's approach here mirrors what we have seen in past technology shifts. The company spends a year filling out every product tier with a clear purpose — in this case, AI performance — and lets the results accumulate. Having watched this pattern across the PC era, the internet buildout, and the smartphone transition, I'd say the strategy is familiar. What is different now is that Apple controls both the chips and the software, which makes it faster to execute but also means outside developers and reviewers are more dependent on Apple's own claims until independent testing catches up.