Mac mini M4 Pro 64 GB
The cheapest Apple machine with enough memory bandwidth to be interesting for local inference. A useful floor for the 'what is the minimum viable machine' question.
ApplesystemAvailable17 models fit0 measured
Memory
64 GB
Official spec LPDDR5X unified
Memory bandwidth
273 GB/s
Official spec ~70% achieved in practice
Specifications
CPU
Apple M4 Pro, 14-core Official spec
GPU
20-core GPU Official spec
Memory
64 GB LPDDR5X unified Official spec
Memory bandwidth
273 GB/s Official spec source
Achieved bandwidth ⓘ
~70% (191 GB/s) Assumption
FP16 compute ⓘ
~17 TFLOPS Estimated
Architecture
Unified memory (CPU and GPU share one pool) Official spec
Nominal power
90 W Official spec
Measured load power
62 W Measured
Idle power
4 W Measured
Released
8 Nov 2024 Official spec
273 GB/s is a hard ceiling: usable for 30B-class MoE models, slow for anything dense above 14B. Its idle power of about 4 W makes it unusually cheap to leave running as an always-on inference box.
Standout models on this machine
Fastest
gpt-oss-20b — ~64.2–92.4 t/s
MXFP4 · estimated confidence
Largest that fits
Compare against
Related hardware
Model performance on Mac mini M4 Pro 64 GB0 measured, 17 estimated
17 of 17 rows
| Model↕ | Quant | Memory↕ | Decode▼ | Prefill↕ | Context | Fit | Confidence↕ |
|---|---|---|---|---|---|---|---|
| gpt-oss-20b OpenAI · 20.915B (3.6B active) | MXFP4recommended | 13.0 GB | ~64.2–92.4 t/s | ~140–291 t/s | 128K | Comfortable | Estimated |
| Nemotron 3.5 Lightning 30B-A3B NVIDIA · 31.578B (3B active) | Q8_0recommended | 33.9 GB | ~39.8–57.2 t/s | ~125–259 t/s | 1024K | Comfortable | Estimated |
| Qwen3-Coder 30B-A3B Alibaba Qwen · 30.532B (3.3B active) | Q8_0recommended | 33.5 GB | ~36.3–52.2 t/s | ~121–252 t/s | 192K | Comfortable | Estimated |
| Qwen3 30B-A3B Alibaba Qwen · 30.532B (3.3B active) | Q8_0recommended | 33.5 GB | ~36.3–52.2 t/s | ~121–252 t/s | 40K | Comfortable | Estimated |
| Gemma 4 26B-A4B Google DeepMind · 25.806B (3.8B active) | Q8_0recommended | 29.7 GB | ~31.7–45.6 t/s | ~123–255 t/s | 96K | Comfortable | Estimated |
| Qwen3 8B Alibaba Qwen · 8.191B | Q8_0recommended | 10.6 GB | ~17.6–25.4 t/s | ~148–308 t/s | 40K | Comfortable | Estimated |
| Phi-4 14B Microsoft · 14.66B | Q8_0recommended | 17.8 GB | ~9.9–14.3 t/s | ~82.9–172 t/s | 16K | Comfortable | Estimated |
| Devstral Small 24B Mistral AI · 23.572B | Q8_0recommended | 26.8 GB | ~6.2–8.9 t/s | ~51.6–107 t/s | 128K | Comfortable | Estimated |
| Mistral Small 3.2 24B Mistral AI · 24.011B | Q8_0recommended | 27.2 GB | ~6.1–8.8 t/s | ~50.6–105 t/s | 128K | Comfortable | Estimated |
| Gemma 3 27B Google DeepMind · 27.432B | Q8_0recommended | 33.4 GB | ~5.3–7.7 t/s | ~44.3–92 t/s | 32K | Comfortable | Estimated |
| Qwen3.8 27B Alibaba Qwen · 27.781B | Q8_0recommended | 30.4 GB | ~5.3–7.6 t/s | ~43.8–90.9 t/s | 256K | Comfortable | Estimated |
| Qwen3.6 27B Alibaba Qwen · 27.781B | Q8_0recommended | 30.4 GB | ~5.3–7.6 t/s | ~43.8–90.9 t/s | 256K | Comfortable | Estimated |
| Granite 4.2 30B IBM · 29.277B | Q8_0recommended | 33.4 GB | ~5–7.2 t/s | ~41.5–86.2 t/s | 64K | Comfortable | Estimated |
| Gemma 4 31B Google DeepMind · 31.273B | Q8_0recommended | 41.0 GB | ~4.7–6.7 t/s | ~38.9–80.7 t/s | 16K | Comfortable | Estimated |
| Qwen3 32B Alibaba Qwen · 32.762B | Q8_0recommended | 37.1 GB | ~4.5–6.4 t/s | ~37.1–77.1 t/s | 40K | Comfortable | Estimated |
| DeepSeek-R1-Distill 32B DeepSeek · 32.764B | Q8_0recommended | 37.1 GB | ~4.5–6.4 t/s | ~37.1–77.1 t/s | 64K | Comfortable | Estimated |
| Llama 3.3 70B Meta · 70.554B | MLX 4-bitrecommended | 43.1 GB | ~3.9–5.7 t/s | ~17.2–35.8 t/s | 32K | Comfortable | Estimated |
Rows in grey are estimates from our bandwidth model, not measurements — they are shown as a range and never as a precise figure. Use the “measured only” filter to see just the 0 pairings on this machine that a real benchmark backs.
Submit a benchmarkContributions are reviewed before publication.
Ownership economicsUnited States (federal) · C corporation · 8h/day
Monthly economic cost
$24
Calculated after tax
Codex/Claude Code
$100/mo
≈ $100/mo · local is $76 less
Net cash at purchase
$2,251
Calculated VAT not reclaimable
Total over 5 years
$1,414
Calculated after tax, after resale
Cost per USD/1M tokens
$2.38
Calculated 9.9M tokens/month
The $100 comparison uses the Codex Pro 5x / Claude Max 5x plans. A fixed planning conversion is used for USD. This compares monthly spend only: subscriptions have usage limits, local hardware has different capabilities and constraints, and taxes or regional pricing may change the charged amount. Prices checked 7 September 2026.
Purchase-price input: $2,251 in United States (federal), tax/VAT excluded · estimated · Apple Store NL (converted catalogue estimate). This is the localized purchase input; the after-tax economic cost is calculated separately below.
Monthly breakdown
| Depreciation 1,576 over 5 years, straight-line to a 675 residual | $26.26 |
| Electricity 2.7 kWh/month at 0.140/kWh | $0.38 |
| Cost of capital 4.0%/yr on 1,463 average capital employed | $4.88 |
| Monthly cost before tax | $31.53 |
| Electricity tax shield Running costs are deductible business expenses | −$0.08 |
| First-year expensing §179 (100.0%) | −$7.88 |
| Monthly economic cost after tax | $23.57 |
Three different numbers, deliberately
Cash cost
$2,251
Money that leaves the bank account on day one, net of reclaimable VAT.
Accounting depreciation
$26.26/month
$1,576 written down over 5 years to a $675 residual.
After-tax economic cost
$23.57/month
Depreciation plus running costs plus cost of capital, less the tax those deductions save. This is the figure to compare between machines.
Assumptions and sources (verified 2026-09-07)
VAT rate
0% Official spec
VAT recoverable
0% Official spec
Federal C-corporation estimate at the flat 21% rate. Pass-through entities should use the sole-proprietor estimate as a rougher proxy.
Effective deduction rate
21.00% Calculated
Headline marginal rate 21.00%.
Depreciation
5 years, straight-line Assumption
Residual value
$675 (30%) Assumption
Extrapolated from the three-year curve.
Electricity
$0.140/kWh Assumption
Average power draw
16 W Calculated
Load 62 W for 20% of powered hours, idle 4 W for the rest — a machine that is on is not generating tokens the whole time.
Investment allowances
§179 100.0% Official spec
Worth $473 in total — first-year expensing that replaces later tax depreciation.
Cost of capital
$4.88/month Assumption
- VAT rate: No US federal VAT · verified 2026-09-07
- Marginal tax rate: IRS Publication 542 — Corporations · verified 2026-09-07
- VAT recoverable fraction: site assumption · verified 2026-09-07
- Useful life: IRS Publication 946 — How To Depreciate Property · verified 2026-09-07
- Electricity price: Site assumption · verified 2026-09-07
Assumes the machine generates tokens 20% of its 176 powered hours per month. Cost per token scales inversely with this number — halve the utilisation and the cost per token doubles.
Federal planning estimate only, not tax advice. State and local income tax, sales/use tax and incentives are excluded. It assumes 100% business use, a Section 179 election, enough business income to use it, and that the full annual limit remains available.