Mixtral 8x22B on Mac Studio M3 Ultra 256 GB

140.63B (39B active) on 256 GB at 819 GB/s. Hardware details · Model details

Yes. Mixtral 8x22B fits on Mac Studio M3 Ultra 256 GB at the recommended Q8_0 configuration, requiring approximately 148.9 GB at 8K context. Practical context capacity is 64K.
CompatibilityComfortable
Calculated fit
Yes
Calculated Uses under 80% of usable memory. Room for a long context and other work.
Recommended quantization
Q8_0
Calculated gguf
Memory required
148.9 GB
Calculated of 217.6 GB usable — 68%
Max practical context
64K
Calculated model supports 64K

Memory budget at 8K context

Model weights ⓘ
141.9 GB Calculated
KV cache ⓘ
1.8 GB Calculated
Runtime overhead ⓘ
5.3 GB Estimated
Total required
148.9 GB Calculated
Headroom ⓘ
68.7 GB Calculated
Highest-precision quantization that leaves headroom: uses 68% of usable memory at 8K context.
Unified memory: weights, KV cache and the OS share the same 256 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 140.63B parameters must be resident in memory even though only ~39B are active per token. Memory follows total parameters; speed follows active parameters.
Performance

Estimated performance · recommended Q8_0

Decode, prefill and TTFT below are estimates for Q8_0. We do not have a comparable Q8_0 measurement on this machine.

Estimated decode
—
Estimated Q8_0; calculated range
Estimated prefill
—
Estimated Q8_0; calculated range
Estimated TTFT at 8K
~41.9–86.8 s
Estimated Q8_0; calculated range
Hardware load reference
195 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
  1. Decode: reading 39.0B active parameters at 8.50 bits/weight takes 71.44 ms at 819 GB/s x 71% achieved efficiency.
  2. MoE routing penalty of 15% applied: expert gathers are less bandwidth-efficient than a dense sweep.
  3. Prefill: 57 TFLOPS (FP16) x 1.72 calibrated against measured prefill on this platform x 22% assumed model-FLOPs utilisation, divided by 2 x 74.1B parameters per token.
  4. Prefill uses 74.1B effective parameters, not the 39B 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
QuantizationResidentDiskTotal neededUtilisationFitMax context
MLX 4-bit76.6 GB76.6 GB81.7 GB38%Comfortable64K
Q4_K_M81.8 GB81.8 GB87.0 GB40%Comfortable64K
Q5_K_M95.9 GB95.9 GB101.6 GB47%Comfortable64K
Q8_0recommended141.9 GB141.9 GB148.9 GB68%Comfortable64K
BF16261.9 GB261.9 GB272.6 GB125%Does not fit0
Cost of running Mixtral 8x22B on this machineUnited States (federal) · C corporation · 8h/day
Monthly economic cost
$60
Calculated after tax
Codex/Claude Code
$100/mo
≈ $100/mo · local is $40 less
Net cash at purchase
$5,674
Calculated VAT not reclaimable
Total over 5 years
$3,575
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: $5,674 in United States (federal), tax/VAT excluded · estimated · Apple Refurbished / used market (converted catalogue estimate). This is the localized purchase input; the after-tax economic cost is calculated separately below.

Monthly breakdown

Depreciation
3,972 over 5 years, straight-line to a 1,702 residual
$66.20
Electricity
8.6 kWh/month at 0.140/kWh
$1.20
Cost of capital
4.0%/yr on 3,688 average capital employed
$12.29
Monthly cost before tax$79.69
Electricity tax shield
Running costs are deductible business expenses
−$0.25
First-year expensing
§179 (100.0%)
−$19.86
Monthly economic cost after tax$59.58

Three different numbers, deliberately

Cash cost
$5,674
Money that leaves the bank account on day one, net of reclaimable VAT.
Accounting depreciation
$66.20/month
$3,972 written down over 5 years to a $1,702 residual.
After-tax economic cost
$59.58/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
$1,702 (30%) Assumption
Extrapolated from the three-year curve.
Electricity
$0.140/kWh Assumption
Average power draw
49 W Calculated
Load 195 W for 20% of powered hours, idle 12 W for the rest — a machine that is on is not generating tokens the whole time.
Investment allowances
§179 100.0% Official spec
Worth $1,192 in total — first-year expensing that replaces later tax depreciation.
Cost of capital
$12.29/month Assumption
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.