Qvq Max
qwen/qvq-max
Qwen (Alibaba)· 128K ctx
open weights quant: unknown
spread (output, p90 ÷ p10)
— too few providers
$4.59 – $4.59 · 1 providers · max÷min 1.0×
1 provider priced · 2 offerings
Cheapest / median / spread use providers that serve the model on their own capacity: first-party lab, hyperscaler, inference host.
Blended $/M assumptions
blended = 90.9% input × [(1 − 70% cache) × input + 70% cache × cache-read] + 9.1% × output + cache writes (30% of uncached input) where a write price is published
A provider that publishes no cache price is blended with cache share 0 (all input at the input price) and labelled “no cache price published” — never penalised against identical in/out prices.
| Provider | Input | Output | Cache rd | Blended |
|---|---|---|---|---|
Alibaba (China)CNfirst-party first-party lab | $1.15 | $4.59 | — | $1.46no cache price published |
NanoGPT router / aggregator | $1.20 | $4.80 | $0.60 | $1.15 |
Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.