Qwen3.8 Flash Next on CPU-only workstation (Ryzen 9950X, 192 GB DDR5)
180B (6B active) on 192 GB at 90 GB/s. Hardware details · Model details
CompatibilityComfortable
Fits
Yes
Calculated Uses under 80% of usable memory. Room for a long context and other work.
Recommended quantization
Q4_K_M
Calculated gguf
Memory required
109.6 GB
Calculated of 176.6 GB usable — 62%
Max practical context
256K
Calculated model supports 256K
Memory budget at 8K context
Model weights ⓘ
104.7 GB Calculated
KV cache ⓘ
0.8 GB Calculated
Runtime overhead ⓘ
4.1 GB Estimated
Total required
109.6 GB Calculated
Usable memory ⓘ
176.6 GB Assumption
Headroom
67.1 GB Calculated
Highest-precision quantization that leaves headroom: uses 62% 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 180B parameters must be resident in memory even though only ~6B are active per token. Memory follows total parameters; speed follows active parameters.
PerformanceEstimated0/10
Decode (generation)
~8.5–12.3 t/s
Estimated calculated, not measured
Prefill (prompt)
~12.7–26.3 t/s
Estimated
TTFT at 8K
~311.5–645.2 s
Estimated time to first token
Power while generating
190 W
Measured 0.16 tokens/s per watt
These figures are estimated, not measured. Nobody has published a benchmark of Qwen3.8 Flash Next on this machine that we know of, so we calculate throughput from memory bandwidth and active parameter count. Estimates are always shown as a range. Expand the working below to see exactly how the number was produced.
How the estimate is calculated
- Decode: reading 6.0B active parameters at 4.85 bits/weight takes 81.19 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 32.9B parameters per token.
- Prefill uses 32.9B effective parameters, not the 6B 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.
This model at other quantizations on this machine
| Quantization | Weights | Total needed | Utilisation | Fit | Max context |
|---|---|---|---|---|---|
MLX 4-bit | 98.1 GB | 102.8 GB | 58% | Comfortable | 256K |
Q4_K_Mrecommended | 104.7 GB | 109.6 GB | 62% | Comfortable | 256K |
Q8_0 | 181.7 GB | 188.9 GB | 107% | Does not fit | 0 |
BF16 | 335.3 GB | 347.1 GB | 196% | Does not fit | 0 |
Cost of running Qwen3.8 Flash Next 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
$15.62
Calculated 1.3M 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.
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
6.8 t/s
Calculated purchase price only
Decode per $100/month
50.5 t/s
Calculated after-tax ownership cost
Tokens per joule
0.16
Calculated same as tokens/s per watt
USD per 1M output tokens
$15.62
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 | $2 | API cheaper | Different model |
| DeepSeek-V4 Flash DeepSeek | $0.66 | 8.5M/mo | $3 | API cheaper | Different model |
| DeepSeek-V4 Pro DeepSeek | $1.98 | 2.8M/mo | $10 | API cheaper | Different model |
| DeepSeek V4 Pro (Together) Together AI | $4.40 | 1.0M/mo | $28 | Local cheaper | Different model |
| GLM-5.3 Z.ai | $4.40 | 1.3M/mo | $21 | About equal | Different model |
| Claude Haiku 4.5 Anthropic | $5.00 | 1.6M/mo | $17 | 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.
Qwen3.8 Flash Next on other hardware
| Hardware | Decode | Price | Fit |
|---|---|---|---|
| Quad RTX 5090 workstation (4x 32 GB) | ~307–442 t/s | €13.999 | Borderline |
| Mac Studio M5 Ultra 256 GB | ~162–233 t/s | €9.599 | Comfortable |
| Mac Studio M5 Ultra 512 GB | ~101–145 t/s | €12.599 | Comfortable |
| Mac Studio M3 Ultra 256 GB | ~97.9–141 t/s | €6.299 | Comfortable |
| Mac Studio M5 Max 128 GB | ~97.9–141 t/s | €4.799 | Borderline |
| Mac Studio M3 Ultra 512 GB | ~58.9–84.7 t/s | €9.199 | Comfortable |
Nearest alternatives to the CPU-only workstation (Ryzen 9950X, 192 GB DDR5)
Qwen3.8 Flash Next on Quad RTX 5090 workstation (4x 32 GB)128 GBQwen3.8 Flash Next on Used RTX 4090 workstation (24 GB)24 GBQwen3.8 Flash Next on Radeon AI PRO R9700 workstation (32 GB)32 GBQwen3.8 Flash Next on RTX PRO 6000 Blackwell workstation (96 GB)96 GBQwen3.8 Flash Next on RTX PRO 6000 Max-Q workstation (96 GB, 300 W)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) →
Page quality score (why this page is or is not indexed)
10/13 — indexed. Generated pages are gated so we do not ask a search engine to rank a page with nothing computed to say. The directive is emitted in the page head via the metadata API, not in the body, so it is authoritative.
- Has a memory-fit calculation (+3)
- Has a performance estimate (+2)
- Has a price (+2)
- Has an ownership economics calculation (+2)
- Well connected (18 internal links) (+1)