Claude Opus 4.6
anthropic/claude-opus-4.6
spread (output, p90 ÷ p10)
1.0×
$25.00 – $30.00 · 11 providers · max÷min 1.2×
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 |
|---|---|---|---|---|
Kilo Gateway router / aggregator | $4.00 | $20.00 | $0.40 | $3.57 |
Poe router / aggregator | $4.30 | $21.00 | $0.43 | $3.79 |
Amazon Bedrock hyperscaler | $5.00 | $25.00 | $0.50 | $4.47 |
ZenMux router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
AIHubMixtiered router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
OrcaRouter router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Perplexity Agent first-party lab | $5.00 | $25.00 | $0.50 | $3.95 |
Pioneer unclassified | $5.00 | $25.00 | $0.50 | $4.47 |
Requesty router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Vercel AI Gateway router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Abacus router / aggregator | $5.00 | $25.00 | — | $6.82no cache price published |
Anthropicfirst-party first-party lab | $5.00 | $25.00 | $0.50 | $4.47 |
Auriko unclassified | $5.00 | $25.00 | $0.50 | $4.47 |
Azuretiered hyperscaler | $5.00 | $25.00 | $0.50 | $4.47 |
Azure Cognitive Servicestiered hyperscaler | $5.00 | $25.00 | $0.50 | $4.47 |
Cloudflare AI Gateway router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Databricks inference host | $5.00 | $25.00 | $0.50 | $4.47 |
Eden AI router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
FreeModel unclassified | $5.00 | $25.00 | $0.50 | $4.47 |
FrogBot unclassified | $5.00 | $25.00 | $0.50 | $4.47 |
GMI Cloud inference host | $5.00 | $25.00 | $0.50 | $3.95 |
Vertextiered hyperscaler | $5.00 | $25.00 | $0.50 | $4.47 |
Vertex (Anthropic)tiered hyperscaler | $5.00 | $25.00 | $0.50 | $4.47 |
Impossibl unclassified | $5.00 | $25.00 | $0.50 | $4.47 |
Jiekou.AI reseller | $5.00 | $25.00 | — | $6.82no cache price published |
LLM Gateway router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Merge Gateway router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
NanoGPT router / aggregator | $5.00 | $25.00 | $0.50 | $3.95 |
NEAR AI Cloud inference host | $5.00 | $25.00 | $0.50 | $4.47 |
Neon router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Ofox reseller | $5.00 | $25.00 | $0.50 | $4.47 |
OpenCode Zen router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
OpenRoutertiered router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Opper router / aggregator | $5.00 | $25.00 | $0.50 | $4.47 |
Cortecs router / aggregator | $5.31 | $26.56 | $0.53 | $4.75 |
Venice AIoutlier inference host | $6.00 | $30.00 | $0.60 | $5.36 |
Subscription plans (flat fee, not per-token) · 1
Flat-fee plan — the per-token price is bundled into a subscription, so $0/token is not free. Excluded from cheapest, median and spread.
- DevPass (LLM Gateway) claude-opus-4-6subscription plan
Data: models.dev (MIT), snapshot 2026-09-20. Prices are vendor-published; verify before purchase. Methodology.