- Estimated cost per request
- $0.002
- Estimated monthly cost
- $20.00
Example rates only. Enter your own prices. Inputs stay in this page; no API key or request is sent.
How to calculate LLM token cost
Subtract cached input from total input, multiply each token category by its rate per million, divide by one million and add the flat request charge. Multiply the result by monthly requests to estimate a month at the same request size.
request cost = ((input − cached) × input rate + cached × cached rate + output × output rate) / 1,000,000 + request charge
What if cached input has no separate price?
Set the cached input price equal to the input price, or enter zero cached tokens. A zero cached-input price means those tokens are free; it does not mean “use the input price.” Total input must include the cached portion to avoid double counting.
Where should I get the rates?
Use your provider’s rate card or the live model catalogue. This calculator uses only the rates you enter and does not fetch or guarantee current provider pricing. Airside providers can use catalogue prices as a starting point when filing eligible flat tariffs.
What does the estimate exclude?
It excludes context tiers, cache-write fees, media-specific rates, time-based pricing, taxes, discounts and gateway margin. It is not a provider payout estimate. For billing units and filing behavior, read the inference pricing guide.