Mercury 2.5
inception/mercury-2.5
Inception· 2026-09-08· 260K ctx
API onlytool callreasoningstructured quant: unknown
spread (output, p90 ÷ p10)
— too few providers
$0.150 – $0.938 · 2 providers · max÷min 6.3×
2 providers 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 |
|---|---|---|---|---|
NanoGPT router / aggregator | $0.0400 | $0.150 | $0.00400 | $0.0271 |
Inceptionfirst-party first-party lab | $0.0400 | $0.150 | $0.00400 | $0.0271 |
OpenRouter router / aggregator | $0.0400 | $0.150 | $0.00400 | $0.0271 |
Vercel AI Gateway router / aggregator | $0.0400 | $0.150 | $0.00400 | $0.0271 |
Kilo Gateway router / aggregator | $0.200 | $0.750 | $0.0200 | $0.135 |
Venice AI inference host | $0.250 | $0.938 | $0.0250 | $0.169 |
AI/ML API unclassified | $0.28 | $1.03 | $0.03 | $0.19 |
Data: models.dev (MIT), snapshot 2026-10-17. Prices are vendor-published; verify before purchase. Methodology.