Qwen3 Max
qwen/qwen3-max
spread (output, p90 ÷ p10)
— too few providers
$6.00 – $8.45 · 3 providers · max÷min 1.4×
Cheapest / median / spread use providers that serve the model on their own capacity: first-party lab, hyperscaler, inference host.
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 |
|---|---|---|---|---|
Ofox reseller | $0.36 | $1.43 | $0.07 | $0.27 |
Merge Gateway router / aggregator | $0.36 | $1.43 | $0.07 | $0.27 |
OrcaRouter router / aggregator | $0.36 | $1.43 | — | $0.46no cache price published |
LLM Gateway router / aggregator | $0.84 | $3.38 | — | $1.08no cache price published |
OpenRoutertiered router / aggregator | $0.78 | $3.90 | $0.16 | $0.75 |
Kilo Gateway router / aggregator | $0.78 | $3.90 | $0.16 | $0.75 |
EmpirioLabs AItiered unclassified | $1.08 | $5.52 | $1.08 | $1.48 |
ZenMux router / aggregator | $1.20 | $6.00 | — | $1.64no cache price published |
Alibabafirst-party first-party lab | $1.20 | $6.00 | — | $1.64no cache price published |
Cloudflare AI Gateway router / aggregator | $1.20 | $6.00 | — | $1.64no cache price published |
Deep Infratiered inference host | $1.20 | $6.00 | $0.24 | $1.03 |
Eden AI router / aggregator | $1.20 | $6.00 | $0.24 | $1.15 |
LLMTR reseller | $1.20 | $6.00 | — | $1.64no cache price published |
Vercel AI Gateway router / aggregator | $1.20 | $6.00 | $0.24 | $1.03 |
Abacus router / aggregator | $1.20 | $6.00 | — | $1.64no cache price published |
NanoGPT router / aggregator | $1.20 | $6.00 | $0.60 | $1.25 |
NovitaAI inference host | $2.11 | $8.45 | — | $2.69no cache price published |
Subscription plans (flat fee, not per-token) · 3
Flat-fee plan — the per-token price is bundled into a subscription, so $0/token is not free. Excluded from cheapest, median and spread.
- Alibaba Coding Plan qwen3-max-2026-01-23subscription plan
- Alibaba Coding Plan (China) qwen3-max-2026-01-23Mainland-China endpointsubscription plan
- DevPass (LLM Gateway) qwen3-maxsubscription plan
Free tiers (rate-limited) · 2
No payment required; rate limits and quotas apply (not tracked yet). Excluded from cheapest, median and spread.
- iFlow qwen3-maxChina-hosted endpoint (iflow.cn)iFlow account + API keyfree tier
- iFlow qwen3-max-previewChina-hosted endpoint (iflow.cn)iFlow account + API keyfree tier
Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.