| L1 instruction cache | L1 data cache | L2 cache | Minimum clock | Maximum clock | |
|---|---|---|---|---|---|
| M6 E-core | 160KB | 128KB | 8MB shared between 6-core cluster | 1,010 MHz | 2,908 MHz |
| M6 P-core | 128KB | 64KB | 20MB split between P- and S-cores | 1,440 MHz | 4,788 MHz |
| M6 Super core | 192KB | 96KB | 20MB split between P- and S-cores | 1,440 MHz | 4,788 MHz |
The M6 gets additional GPU cores for the first time since the M2, increasing the count from 10 cores to 12. These cores include improved versions of the integrated neural accelerators Apple introduced in the M5-series GPU for speeding up AI and machine-learning workloads like ray-tracing and MetalFX upscaling. A “Dual 16-core Neural Engine” also bumps the number of Neural Engine cores from 16 to 32. Saying you have 2×16 of something strikes me as a roundabout way of saying you have 32 of something, but I suppose the end result is the same either way.
One small wrinkle in the M6 variants the Mac mini uses: The basic 16GB Mac mini offers 153 GB/s of memory bandwidth, the same amount as the M5. The 24GB and 32GB versions offer 170 GB/s, an 11 percent increase that should marginally improve graphics performance and the speed of any local AI models you run on this machine. Our review unit is the 24GB version.
What of the M6 Pro, M6 Max, or M6 Ultra? If the rumors are true, the M6 could end up being a lonely chip.
Bloomberg’s Mark Gurman reports that Apple isn’t planning a full M6 chip generation and that it could skip directly to M7 branding for the Pro, Max, and Ultra chips. That would be a first for the company, which skipped shipping an M4 Ultra but has otherwise shipped a baseline, Pro, and Max-series chip in each generation so far.
But it makes sense for a couple of reasons. The first is that most of the M6’s gains can be attributed to its extra CPU and GPU cores relative to the M5, and the M5 Pro, M5 Max, and M5 Ultra all obviously still have more of both kinds of cores than the M6 does. The second is that it’s still pretty early for TSMC’s 2 nm process, following three generations where Apple used versions of TSMC’s 3 nm process, and Apple is probably fighting with every other fabless chipmaker on Earth to reserve capacity. Apple could be sticking with smaller, simpler 2 nm chips for now and saving its bigger and more complex chips for later, when capacity will be higher, yields will be better, and costs may be lower.