LLM API cost calculator

Estimate what an AI feature will actually cost to run. Set the tokens per request and your volume to see cost per call, per day, per month and per year — then compare models across providers to find the cheapest one that does the job.

Your usage

Reused prompt prefix, billed at ~25% (rough).

Indicative list prices in USD, last reviewed . Providers change pricing and offer batch/cached discounts — confirm on the provider's page before budgeting.

Estimated cost

Cost per month
Per request
input + output
Per day
requests/day
Per year
at this volume
Per 1K requests
handy unit

Compare every model at this usage

Monthly cost for the same tokens and volume. Cheapest first; your pick is highlighted.

Output usually dominates

Output tokens are typically 3–5× the input price. A verbose model that writes long answers can cost more than a pricier model told to be terse. Cap output length first.

Right-size the model

A small model at a tenth of the price often handles classification, extraction, and routing fine. Reserve the flagship for the calls that genuinely need it.

Caching and batch cut bills

Reused prompt prefixes (cached input) and asynchronous batch endpoints are heavily discounted by most providers — worth it once volume is real.

About LLM API costs

How is LLM API cost calculated?

Cost per request = (input tokens ÷ 1,000,000 × input price) + (output tokens ÷ 1,000,000 × output price). Multiply by requests per day for daily spend, then by 30.44 for a month. Input and output are priced separately, and providers quote per million tokens.

What is a token?

A token is a chunk of text — roughly 4 characters or 0.75 words in English. Both your prompt (input) and the model's reply (output) count.

Why is output more expensive than input?

Generating tokens costs more compute than reading them, so output is usually priced 3–5× higher. Long completions can dominate your bill — trimming output length is often the fastest saving.

Are these prices current?

They're indicative list prices, dated in the tool. Providers change pricing and offer batch, cached, and committed-use discounts. Always confirm on the provider's pricing page before budgeting.

Is my data stored anywhere?

No. This calculator runs entirely in your browser. Nothing you enter is sent to any server.

More tools for AI builders

Get new tools as they ship

Get an email when we ship the next AI cost or unit-economics tool.

No spam, no signup needed to use any tool. Unsubscribe any time.
Thanks — you're on the list. We'll only email when we ship.