LLM Token Cost Calculator
Estimate input, cached-input and output token charges for an entered request volume.
Your result
EVERYDAYEnter your values and select calculate.
On this page How it works & practical examples
How to use the LLM Token Cost Calculator
Supply the applicable provider rates and average token counts. Cached tokens are part of the total input, not extra tokens. Keep model, token-accounting and currency definitions consistent when estimating a batch.
The calculation
Worked example
1,000 input tokens including 200 cached, 500 output tokens and example rates of $1, $0.25 and $2 per million cost $0.00185 per request, or $18.50 for 10,000 requests.
Assumptions and limits
Example rates are editable illustrations, not current provider pricing. Tool calls, images, storage, reasoning-token definitions, free tiers, taxes and other charges are not inferred.
Common questions
Are these live model prices?
No. Enter the rates from your provider’s applicable pricing.
Do cached tokens count twice?
No. They are subtracted from uncached input before the separate cache rate is applied.