we_are_coded.by CODE · The world, decoded
БГ

Apple · Chips

Apple put M6 in the MacBook Pro and iMac: by Apple's own numbers, up to 25% faster AI than M5 on the laptop and up to 5x faster than M4 on the iMac

On 27 October Apple announced a 14-inch MacBook Pro and a new iMac with the M6 chip and presents them as platforms for AI on the machine itself. By Apple's figures, AI workloads on the laptop are up to 25% faster than M5, and the iMac is up to 5x faster than M4. Orders open 28 October, stores from 6 November.

Source
Apple Newsroom
Published
The event
Checked on
Responsible editor
Cvetelin Ivanov
How we work
MethodCorrections

The laptop is up to 25% faster at AI. The iMac is up to 5x faster. Both numbers are Apple's and both are for the same chip, but they do not start from the same place: the laptop is compared with M5, the iMac with M4.

Fact

The facts: on 27 October 2026 Apple announced a 14-inch MacBook Pro with the M6 chip and a new iMac with M6. By Apple's figures, M6 in the MacBook Pro has a 12-core CPU, a 12-core GPU with Neural Accelerators in each core and a Dual 16-core Neural Engine; memory is 16GB, configurable up to 32GB, with bandwidth up to 170GB/s. According to Apple, the GPU is up to 30% faster at graphics and up to 25% faster at AI workloads compared with M5, and the Neural Engine is up to 1.9x faster than M5. Apple calls the CPU core the fastest in the world - that is its own claim. The iMac is up to 5x faster at on-device AI than M4, and in LM Studio prompt processing is up to 12x faster than on an iMac with M1 and up to 5x faster than on M4 (by Apple's figures). The 14-inch MacBook Pro starts at $1,999 in the US, or $1,899 with education savings. Both computers can be pre-ordered from 28 October and reach stores on Friday, 6 November. Orders open in 35 countries and regions, including the US.

Apple sells the chip with one argument, and it sits in a sentence about the laptop: M6 is a platform for AI, including “running LLMs locally, all while preserving user privacy”. A model that lives on your machine is a strong argument. But the main text of the announcements names no model it was measured with, and no amount of memory it needed.

Every number with up to in front of it has a baseline, and Apple picks the baseline.

The laptop announcement carries a second number for the same kind of work: up to 1.3x faster time to first token in LLM prompt processing than M5. That is about 30%, while the headline figure is 25%. It may be rounding, it may be a different test; the announcement does not say. Both start with up to, which means the best result of their own test. Yours may be lower.

There are no independent tests yet. The first machines reach people on 6 November, and from then on Apple's percentages will have something to be compared with.

The visual is generated code-art, no third-party images.