Step 3.7 Flash 198B-A11B on Mac mini M4 Pro 64 GB
201.365B (11B active) on 64 GB at 273 GB/s. Hardware details · Model details
No calculated fit. Step 3.7 Flash 198B-A11B does not fit under standard overhead assumptions at
MLX 4-bit: the model calculates 125.2 GB required against 54.4 GB usable.CompatibilityDoes not fit
Calculated fit
No
Calculated Requires more than 97% of usable memory. Needs a smaller quantization or more memory.
Smallest catalogued quantization
MLX 4-bit
Calculated mlx
Memory required
125.2 GB
Calculated of 54.4 GB usable — 230%
Max practical context
0
Calculated model supports 256K
Memory budget at 8K context
Model weights ⓘ
109.7 GB Calculated
KV cache ⓘ
11.3 GB Calculated
Runtime overhead ⓘ
4.3 GB Estimated
Total required
125.2 GB Calculated
Headroom ⓘ
-70.8 GB Calculated
No quantization in our catalogue fits this machine.
Unified memory: weights, KV cache and the OS share the same 64 GB pool. We assume 85% is addressable by the GPU, which matches a headless machine with the wired-memory limit raised.
Mixture of experts: all 201.365B parameters must be resident in memory even though only ~11B are active per token. Memory follows total parameters; speed follows active parameters.
Performance
Estimated performance · recommended MLX 4-bit
Decode, prefill and TTFT below are estimates for MLX 4-bit. We do not have a comparable MLX 4-bit measurement on this machine.
Estimated decode
—
Estimated MLX 4-bit; calculated range
Estimated prefill
—
Estimated MLX 4-bit; calculated range
Estimated TTFT at 8K
~152.9–317.6 s
Estimated MLX 4-bit; calculated range
Hardware load reference
62 W
Measured machine-level load; not this model run
No throughput figures for this configuration. We do not estimate performance for a model that cannot load.
How the estimate is calculated
- Decode: reading 11.0B active parameters at 4.50 bits/weight takes 32.38 ms at 273 GB/s x 70% achieved efficiency.
- MoE routing penalty of 15% applied: expert gathers are less bandwidth-efficient than a dense sweep.
- Prefill: 17 TFLOPS (FP16) x 22% assumed model-FLOPs utilisation, divided by 2 x 47.1B parameters per token.
- Prefill uses 47.1B effective parameters, not the 11B active in decode: a batch of hundreds of tokens routes across most of the expert pool.
- MoE decode depends on how well the engine batches expert gathers; real results vary more than for dense models.
- Prefill throughput is highly engine-dependent. Flash attention, batch size and quantized KV all move this number substantially, and for sparse mixture-of-experts models it is the least reliable figure we produce.
- This is a calculated estimate, not a measurement. It assumes a single request, a short prompt, no speculative decoding and a warm model already resident in memory.
Estimator version estimator_v1. Stored with every estimated row so old estimates can be regenerated when the model improves.
Submit a benchmarkSubmissions stay pending until reviewed.
This model at other quantizations on this machine
| Quantization | Resident | Disk | Total needed | Utilisation | Fit | Max context |
|---|---|---|---|---|---|---|
MLX 4-bitsmallest listed | 109.7 GB | 109.7 GB | 125.2 GB | 230% | Does not fit | 0 |
Q4_K_M | 117.1 GB | 117.1 GB | 132.9 GB | 244% | Does not fit | 0 |
Q5_K_M | 137.4 GB | 137.4 GB | 153.8 GB | 283% | Does not fit | 0 |
Q8_0 | 203.2 GB | 203.2 GB | 221.6 GB | 407% | Does not fit | 0 |
BF16 | 375.1 GB | 375.1 GB | 398.6 GB | 733% | Does not fit | 0 |
Cost of running Step 3.7 Flash 198B-A11B on this machineUnited 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
—
Calculated
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
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.
Step 3.7 Flash 198B-A11B on other hardware
| Hardware | Decode | Price | Fit |
|---|---|---|---|
| NVIDIA DGX H200 (8x H200, 1,128 GB) | ~431–620 t/s | $359,430 * | Comfortable |
| Dell Precision 7960 Rack (2x RTX PRO 6000, 192 GB) | ~143–205 t/s | $36,663 * | Comfortable |
| Lenovo ThinkStation PX (4x RTX PRO 6000, 384 GB) | ~126–181 t/s | $58,553 * | Comfortable |
| Mac Studio M5 Ultra 256 GB | ~84.1–121 t/s | $8,647 * | Comfortable |
| NVIDIA DGX Station GB300 (748 GB) | ~58.9–84.8 t/s | $90,082 * | Comfortable |
| Mac Studio M5 Ultra 512 GB | ~58.3–83.8 t/s | $11,350 * | Comfortable |
Nearest alternatives to the Mac mini M4 Pro 64 GB
Step 3.7 Flash 198B-A11B on Mac Studio M5 Max 64 GB64 GBStep 3.7 Flash 198B-A11B on Mac Studio M5 Max 36 GB36 GBStep 3.7 Flash 198B-A11B on Mac Studio M5 Ultra 96 GB96 GBStep 3.7 Flash 198B-A11B on Mac Studio M5 Max 128 GB128 GBStep 3.7 Flash 198B-A11B on Mac Studio M3 Ultra 256 GB256 GB
Compare Mac mini M4 Pro 64 GB against Mac Studio M5 Max 64 GB →
Compare Mac mini M4 Pro 64 GB against Mac Studio M5 Max 64 GB →