Mistral Small 4 119B-A6.5B on CPU-only workstation (Ryzen 9950X, 192 GB DDR5)
119.401B (6.5B active) on 192 GB at 90 GB/s. Hardware details · Model details
Yes. Mistral Small 4 119B-A6.5B fits on CPU-only workstation (Ryzen 9950X, 192 GB DDR5) at the recommended
Q8_0 configuration, requiring approximately 125.3 GB at 8K context. Practical context capacity is 256K. Expected decode for the recommended configuration is ~4.5–6.5 t/s. EstimatedCompatibilityComfortable
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
125.3 GB
Calculated of 176.6 GB usable — 71%
Max practical context
256K
Calculated model supports 256K
Memory budget at 8K context
Model weights ⓘ
120.5 GB Calculated
KV cache ⓘ
0.2 GB Calculated
Runtime overhead ⓘ
4.6 GB Estimated
Total required
125.3 GB Calculated
Headroom ⓘ
51.3 GB Calculated
Highest-precision quantization that leaves headroom: uses 71% of usable memory at 8K context.
Discrete GPU: 192 GB of VRAM, of which we assume 92% is usable after driver and context overhead.
Mixture of experts: all 119.401B parameters must be resident in memory even though only ~6.5B 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
~4.5–6.5 t/s
Estimated Q8_0; calculated range
Estimated prefill
~14.9–31 t/s
Estimated Q8_0; calculated range
Estimated TTFT at 8K
~264.3–549.9 s
Estimated Q8_0; calculated range
Hardware load reference
190 W
Measured machine-level load; not this model run
How the estimate is calculated
- Decode: reading 6.5B active parameters at 8.50 bits/weight takes 154.16 ms at 90 GB/s x 50% achieved efficiency.
- MoE routing penalty of 15% applied: expert gathers are less bandwidth-efficient than a dense sweep.
- Prefill: 4 TFLOPS (FP16) x 32% assumed model-FLOPs utilisation, divided by 2 x 27.9B parameters per token.
- Prefill uses 27.9B effective parameters, not the 6.5B 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 |
|---|---|---|---|---|---|---|
Q4_K_M | 69.4 GB | 69.4 GB | 72.7 GB | 41% | Comfortable | 256K |
Q5_K_M | 81.5 GB | 81.5 GB | 85.1 GB | 48% | Comfortable | 256K |
Q8_0recommended | 120.5 GB | 120.5 GB | 125.3 GB | 71% | Comfortable | 256K |
Cost of running Mistral Small 4 119B-A6.5B on this machineUnited States (federal) · C corporation · 8h/day
Monthly economic cost
$21
Calculated after tax
Codex/Claude Code
$100/mo
≈ $100/mo · local is $79 less
Net cash at purchase
$1,531
Calculated VAT not reclaimable
Total over 5 years
$1,233
Calculated after tax, after resale
Cost per USD/1M tokens
$29.53
Calculated 0.7M 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: $1,531 in United States (federal), tax/VAT excluded · estimated · NL retail (system build) (converted catalogue estimate). This is the localized purchase input; the after-tax economic cost is calculated separately below.
Monthly breakdown
| Depreciation 1,301 over 5 years, straight-line to a 230 residual | $21.68 |
| Electricity 11.6 kWh/month at 0.140/kWh | $1.63 |
| Cost of capital 4.0%/yr on 880 average capital employed | $2.93 |
| Monthly cost before tax | $26.24 |
| Electricity tax shield Running costs are deductible business expenses | −$0.34 |
| First-year expensing §179 (100.0%) | −$5.36 |
| Monthly economic cost after tax | $20.54 |
Three different numbers, deliberately
Cash cost
$1,531
Money that leaves the bank account on day one, net of reclaimable VAT.
Accounting depreciation
$21.68/month
$1,301 written down over 5 years to a $230 residual.
After-tax economic cost
$20.54/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
$230 (15%) Assumption
Two GPU generations later, the card is worth a fraction of its list price.
Electricity
$0.140/kWh Assumption
Average power draw
66 W Calculated
Load 190 W for 20% of powered hours, idle 35 W for the rest — a machine that is on is not generating tokens the whole time.
Investment allowances
§179 100.0% Official spec
Worth $321 in total — first-year expensing that replaces later tax depreciation.
Cost of capital
$2.93/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.
Efficiency metrics
Decode per $1,000 spent
3.6 t/s
Calculated purchase price only
Decode per $100/month
26.7 t/s
Calculated after-tax ownership cost
Tokens per joule
0.08
Calculated same as tokens/s per watt
USD per 1M output tokens
$29.53
Calculated at 20% utilisation
Local versus hosted APIs
| Hosted model | USD/1M output | Break-even | API at your volume | Verdict | Comparison type |
|---|---|---|---|---|---|
| Qwen3.8 Flash Alibaba Cloud | $0.42 | 13.3M/mo | $1 | API cheaper | Different model |
| DeepSeek-V4.1 Flash DeepSeek | $0.60 | 11.4M/mo | $1 | API cheaper | Different model |
| DeepSeek-V4 Pro DeepSeek | $1.98 | 2.8M/mo | $5 | API cheaper | Different model |
| DeepSeek V4 Pro (Together) Together AI | $4.40 | 1.0M/mo | $15 | API cheaper | Different model |
| GLM-5.3 Z.ai | $4.40 | 1.3M/mo | $11 | API cheaper | Different model |
| Claude Haiku 4.5 Anthropic | $5.00 | 1.6M/mo | $9 | API cheaper | Different model |
Different-model comparison: the hosted model is not the model you would run locally. Treat this as a workload-quality trade-off, not a direct economic equivalence — the frontier model may complete a task in fewer tokens, or complete tasks the local model cannot. Break-even is the monthly output volume at which API spend equals the $21/month economic cost of owning this machine, assuming 8 input tokens per output token. Hosted prices are published in USD. This table is the canonical US default.
Mistral Small 4 119B-A6.5B on other hardware
| Hardware | Decode | Price | Fit |
|---|---|---|---|
| NVIDIA DGX H200 (8x H200, 1,128 GB) | ~483–695 t/s | $359,430 * | Comfortable |
| Quad RTX 5090 workstation (4x 32 GB) | ~261–375 t/s | $12,611 * | Comfortable |
| RTX PRO 6000 Blackwell workstation (96 GB) | ~192–276 t/s | $10,359 * | Fits |
| RTX PRO 6000 Max-Q workstation (96 GB, 300 W) | ~192–276 t/s | $10,809 * | Fits |
| Lenovo ThinkStation PX (4x RTX PRO 6000, 384 GB) | ~185–266 t/s | $58,553 * | Comfortable |
| Mac Studio M5 Ultra 96 GB | ~152–218 t/s | $5,945 * | Fits |
Nearest alternatives to the CPU-only workstation (Ryzen 9950X, 192 GB DDR5)
Mistral Small 4 119B-A6.5B on Quad RTX 5090 workstation (4x 32 GB)128 GBMistral Small 4 119B-A6.5B on RTX 3090 workstation (Ryzen 9 7950X, 64 GB)24 GBMistral Small 4 119B-A6.5B on Used RTX 4090 workstation (24 GB)24 GBMistral Small 4 119B-A6.5B on Radeon AI PRO R9700 workstation (32 GB)32 GBMistral Small 4 119B-A6.5B on RTX PRO 6000 Blackwell workstation (96 GB)96 GB
Compare CPU-only workstation (Ryzen 9950X, 192 GB DDR5) against Quad RTX 5090 workstation (4x 32 GB) →
Compare CPU-only workstation (Ryzen 9950X, 192 GB DDR5) against Quad RTX 5090 workstation (4x 32 GB) →