ailocalinference

← All hardware

MacBook Pro M5 Pro

Laptop · 18-core CPU / 20-core GPU config

AppleLaptop$2,760 · Aug 31, 2026View on Amazonpaid link

Memory
64 GB~48 GB usable
Bandwidth
307GB/s — sets decode speed
Compute
15FP16 TFLOPS — sets prefill
Power
90 Wunder sustained load
Noise
45 dBAyou will hear it
Price per GB
$43of accelerator memory

Price history

tracking

Tracking since 2026-08-31 — not enough history for a trend yet.

1 observation recorded · Last checked 2026-08-31. Check the retailer before acting on this — we would rather say the data is old than pretend otherwise.

What it can run

3 of 6 reference workloads fit. The largest is 70B dense · Q4 (strong general), at an estimated 5 tok/s of single-stream decode — #5 of the 7 machines here that can run it.

WorkloadNeedsFitsPrefill t/sDecode t/sBatched t/s$ per t/s
8B dense · Q4 (chat, drafts)
dense · 4.7 GB read per token
5.4 GBYes14142101$66
32B dense · Q4 (coding)
dense · 19 GB read per token
21.8 GBYes351125$251
70B dense · Q4 (strong general)
dense · 40 GB read per token
46 GBYes16512$552
120B MoE A5B · MXFP4 (gpt-oss class)
MoE · 2.9 GB read per token
72.4 GBNo22144150—
320B MoE A18B · FP8 (frontier-ish)
MoE · 18 GB read per token
368 GBNo63736—
671B MoE A37B · Q4 (DeepSeek class)
MoE · 18.5 GB read per token
437 GBNo30721—

Figures are modelled from this machine's bandwidth and compute, not measured on one. How they are derived →

Alternatives that also run 70B dense · Q4 (strong general)

MachinePriceMemoryDecode t/s$ per t/s
$4,500128 GB10$450
$15,49996 GB29$534
MacBook Pro M5 Pro — this machine
$2,76064 GB5$552
$13,500512 GB20$675
$9,499512 GB13$731
$3,999128 GB4$1,000
$5,199128 GB4$1,300