Planning tool

LLM cost estimator

Model token spending using your own provider rates.

Calculations stay in your browser. Read the privacy notice.

Example: token costs only

10,000 requests × (1,500 input tokens × $1 + 500 output tokens × $4) ÷ 1,000,000 gives $35 per month, or $420 annualized, at these illustrative rates.

Read the result

Replace both rates with the actual provider/model prices and test real token usage. Doubling output tokens to 1,000 raises this example to $55 per month; retries, tools and hosting remain additional.

How it works

Monthly cost = requests × (input tokens × input rate + output tokens × output rate) ÷ 1,000,000. Example rates are placeholders, not a provider quote. Excludes caching discounts, tool calls, retries, hosting, and other fees.

These are planning estimates based on your inputs, not promised savings or a forecast of actual results. Amounts use USD.

All calculators ↗Discuss your assumptions ↗