What can your Mac run?

Choose memory for capacity; check tests for speed.

More unified memory can fit a larger checkpoint or longer context. A faster chip does not make a model fit into less memory. Compare exact GPU core counts, bandwidth, cooling and runtime before treating two Macs as interchangeable.

The current selection covers Mac mini, Mac Studio, MacBook Air and MacBook Pro, with two previous-generation Mac mini variants for existing owners. It is a reviewed selection, not every Apple configuration.

Unified memory needs its own budget.

These Macs contain an integrated Apple GPU. CPU, GPU, macOS and other apps share one memory pool. We reserve 25% (at least 4 GiB) for the system and other apps, then add exact Q4_K_M weight-file bytes, FP16 KV cache and 2 GiB for the inference runtime. The system allowance is our editable planning assumption, not a measured free-memory value or a universal macOS GPU limit.

Nominal Apple GB labels are treated as GiB for this scenario. Context includes input and output tokens; one request, weights kept in memory. Passing the estimate is a shortlist, not a verified deployment. The four-model sample stops at 70B; a 70B result is not a ceiling on a large Mac.

Actual Metal working-set limits, prefill workspace, app usage, runtime allocations and model support may change the result. Swapping to SSD can make a model load while becoming too slow for the intended task. Memory bandwidth alone cannot establish output speed.

Buy, rent or use an API?

Mac purchase references cover the whole computer. Discrete-GPU prices may cover the card only; add the host system before comparing. We have no matching Mac cloud-compute quote yet. Apple hardware financing or a monthly equipment lease is not an hourly inference service.