Pricing
Prices are quoted per million tokens, separately for input and output. In the
API and headers, prices are expressed in USDC microunits (1 USDC =
1,000,000 microunits) — so 400000 means $0.40 per million tokens.
The cap-at-centralized rule
Section titled “The cap-at-centralized rule”Marketplace and key relay listings are capped at the cheapest centralized price for the same model, on both the input and output axes. A provider can list below the centralized price to win traffic, but never above it. Consequences:
- The marketplace is never more expensive than the centralized fallback.
- In auto routing, an over-priced listing simply loses to a cheaper option or to centralized.
- In marketplace-only routing, if every listing exceeds your ceiling, the request returns a no-provider-at-price result instead of overcharging.
Fee split
Section titled “Fee split”Marketplace routes split each settled inference:
| Party | Share |
|---|---|
| Provider | 90% |
| Treasury | 10% |
Users are billed at the operator’s listed (capped) price; the split is applied to that amount at settlement.
Billing basis
Section titled “Billing basis”Billing is based on actual token usage extracted from the response stream, at the price of the provider that served the request.
- Centralized: the serving provider’s own published rates, including its cache-read and cache-write rates. OpenRouter-served requests bill at the cost OpenRouter reports for that request. If a request fails over to a pricier provider, it bills at that provider’s rates, which is what you would pay that provider directly.
- Marketplace and key relays: the listing’s rates. Cache reads and writes bill at the listing’s own cache rates. A listing without them bills cache tokens at the cheapest cache rate any centralized provider publishes for the model, never above the listing’s input rate, or at the input rate when no provider publishes one.
A missing cache price is never treated as free.