Apple M6 Chip Is Coming This Year But It Will Be Unlike Any Apple Silicon Before It

Apple’s M6 chip is arriving later this year with 200 GB/s memory bandwidth, 12 GPU cores, and improved on-device AI performance but without any Pro, Max, or Ultra variants, making it the most strategically unusual Apple Silicon release since the M1 launched the era in 2020.

Apple Silicon has followed a predictable pattern since its debut. A base chip arrives first, powerful, efficient, and impressive. Then the Pro and Max variants follow, offering dramatically more performance for creative professionals and power users. Then the Ultra, connecting two Max chips for workstation-class capability. The cycle repeats with the next generation number. It has been one of the most consistent product cadences in Apple’s hardware history.

The M6 breaks that pattern entirely.

What the M6 Brings

According to Bloomberg, the M6 chip delivers several meaningful upgrades over the current M5 generation.

The headline specification is memory bandwidth. The M5 currently maxes out at 153 gigabytes per second. The M6 pushes that figure to 200 gigabytes per second, a 31 percent increase that has direct implications for on-device AI performance, which is heavily dependent on how quickly the processor can move data between memory and compute units. As Apple Intelligence features become more central to the Mac and iPhone experience, memory bandwidth has become one of the most practically important specifications in Apple Silicon.

The GPU receives a notable upgrade as well. The base M5 chip supports up to 10 GPU cores. The M6 steps up to 12 graphics cores, a 20 percent increase that translates into smoother gaming performance, faster video rendering, and better performance in GPU-accelerated creative and scientific applications. For video editors, 3D artists, and developers working with GPU-intensive workloads, this is a meaningful generational step.

Performance improvements across all CPU cores round out the picture, along with enhancements to video encoding and decoding, areas that Apple has consistently improved with each chip generation, benefiting anyone working with high-resolution video content.

The Decision That Makes M6 Unprecedented

None of the above is particularly surprising for an Apple Silicon generation. What is unprecedented is what Apple will not be releasing alongside the base M6.

Since the M1, every Apple Silicon generation has included Pro and Max variants, chips that multiply the core counts, memory bandwidth, and capability of the base chip for users who need more. The M1, M2, and M3 generations also included Ultra chips, connecting two Max dies for the highest-end Mac Pro and Mac Studio configurations.

The M6 generation will have none of those variants. According to Bloomberg’s Mark Gurman, Apple will release only the base M6. No M6 Pro. No M6 Max. No M6 Ultra.

The reason, according to reporting, is that Apple has determined the M7’s improvements are significant enough to justify skipping the higher-end M6 lineup entirely. The M7 chip, expected in the first half of 2027, is reported to offer substantially bigger upgrades to on-device AI processing, including support for up to 240 gigabytes per second of memory bandwidth. If the M7 is coming relatively quickly and brings substantially better AI capability, releasing M6 Pro and Max variants would create a product generation that professional users would immediately want to skip in favour of M7-based machines.

From Apple’s perspective, the strategic calculus is straightforward: release the base M6 to give the entry-level MacBook Pro a generational update, skip the Pro and Max variants that would cannibalise M7 interest, and let the M7 deliver the full-lineup upgrade that power users are waiting for.

What Products Will Get M6

The M6 chip has been tested in an updated version of the base model MacBook Pro, which currently runs the M5. That device is the most likely recipient of the M6 when it launches later this year.

Whether the M6 reaches any other Apple products, the MacBook Air, Mac mini, or iPad Pro, remains unconfirmed. Given that the M7 is expected in the first half of 2027, Apple has a relatively short window in which M6 makes strategic sense as a product differentiator. The most probable outcome is that the base MacBook Pro receives the M6 upgrade while most of Apple’s other Mac and iPad lines either remain on M5 or wait for the M7.

For the Mac Pro and Mac Studio, which have historically depended on the Max and Ultra variants, there is no M6 path at all. These products will either continue with current M5 Ultra configurations or wait for M7 Ultra and M7 Max variants that arrive after the base M7 launches in 2027.

The AI Angle

The emphasis on memory bandwidth as a primary upgrade metric for the M6 is not coincidental. On-device AI inference, running large language models, image generation, and intelligent feature processing locally rather than in the cloud, is profoundly memory-bandwidth-limited. The faster a chip can move data between its compute units and its unified memory, the faster it can process AI workloads.

Apple’s M5 at 153 GB/s already leads most competing architectures. The M6 at 200 GB/s extends that lead, while the M7 at 240 GB/s will extend it further. Each generation, Apple is building a chip that can run more capable AI models faster, enabling Apple Intelligence features that would require cloud processing on less capable hardware.

For Mac users, this trajectory means each generation of Apple Silicon brings meaningfully better on-device AI capability, smarter writing tools, faster image processing, more capable Siri interactions, and, eventually, the ability to run sophisticated AI models that today require a data centre connection.

What This Means for Buyers

If you are currently on an M3 or M4 MacBook Pro and considering an upgrade, the M6’s release later this year presents a choice point. The base M6 MacBook Pro will offer meaningfully better AI performance, a 12-core GPU, and 200 GB/s memory bandwidth, a genuine step up from M4.

If you are a power user whose work depends on the higher-end chip configurations, video production at scale, 3D rendering, and software compilation, the M6 generation offers you nothing new. Your upgrade path runs directly to the M7 in 2027, which will bring the Pro, Max, and presumably Ultra variants with the full AI performance uplift Apple is building toward.

If you are currently on an M5 base MacBook Pro, the M6 upgrade is more modest. The GPU core increase and memory bandwidth improvement are real, but the generational gap from M5 to M6 in the base configuration is smaller than the gap from M4 or M3. Waiting for M7 is a reasonable choice.

The Bottom Line

The Apple M6 is a chip that tells you as much about what Apple is planning for M7 as it does about its own capabilities. The decision to skip Pro, Max, and Ultra variants is a strategic signal that the M7 generation will be significant, significant enough that Apple does not want to spend 2026 selling M6 Pro machines that power users will immediately wish were M7 Pro machines. The base M6 is a solid generational step for entry-level Mac buyers. For everyone else, the real story begins in 2027.