Every MacBook Pro 16" generation, with the memory bandwidth and unified memory that decide which models it can run. Pick a chip for its full model fit breakdown.
The M5 Max is the current generation at 614 GB/s and up to 128 GB. On the used market the M4 Max is the one to look for: 546 GB/s and up to 128 GB, which is the pairing that decides what runs, and neither figure changes with the age of the machine.
460 to 614 GB/s · 36 to 128 GB unified memory
Holds about a 172B model at Q4
307 GB/s · 24 to 64 GB unified memory
Holds about a 85B model at Q4
410 to 546 GB/s · 36 to 128 GB unified memory
Holds about a 172B model at Q4 · Used market
273 GB/s · 24 to 48 GB unified memory
Holds about a 63B model at Q4 · Used market
300 to 400 GB/s · 36 to 128 GB unified memory
Holds about a 172B model at Q4 · Used market
150 GB/s · 18 to 36 GB unified memory
Holds about a 47B model at Q4 · Used market
400 GB/s · 32 to 96 GB unified memory
Holds about a 129B model at Q4 · Used market
200 GB/s · 16 to 32 GB unified memory
Holds about a 42B model at Q4 · Used market
400 GB/s · 32 to 64 GB unified memory
Holds about a 85B model at Q4 · Used market
200 GB/s · 16 to 32 GB unified memory
Holds about a 42B model at Q4 · Used market
Apple has not announced any of this. The M7 Max and Ultra are the first parts Apple designed after cancelling a generation to reach them, so extrapolating from the M5 under-represents them. These rows assume LPDDR6, whose wider channels grow every bus by half, at its top speed bin by the time the Max and Ultra ship. The 1.5 TB Ultra ceiling is Bloomberg's reported design target, and whether that configuration ships depends on the memory market. Stacked memory or a new package fabric would land above these numbers; nobody outside Apple can price that yet.
Projected chip · expected 2027
691 GB/s · 24 to 96 GB unified memory
18-core CPU · 20 or 24-core GPU · 76 TOPS Neural Engine
Would hold about a 129B model at Q4
Projected chip · expected 2027
1382 GB/s · 48 to 384 GB unified memory
20-core CPU · 56 or 64-core GPU · 76 TOPS Neural Engine
Would hold about a 521B model at Q4
Start from the model and see every Mac that runs it, with fit ratings and speed estimates.
Browse the model database