Home › AI Tools

AI API Cost Calculator

Work out what an AI model API will cost per day, month and year from tokens per request and the price per million tokens.

How it works

AI providers charge separately for input tokens (what you send, including instructions and earlier messages) and output tokens (what the model writes back), usually as a price per million tokens. Cost per request = input tokens × input price ÷ 1,000,000 + output tokens × output price ÷ 1,000,000. Multiply by requests per day and days of use per month. Copy current prices from your provider's official pricing page, since they change often.

Example

Input tokens per request: 1500 · Output tokens per request: 500 · Requests per day: 1000 · Input price ($ per 1M tokens): 2 · Output price ($ per 1M tokens): 10 · Days of use per month: 30 → Monthly cost: $240.00

Formula

per request = in × price in ÷ 1M + out × price out ÷ 1M      month = per request × requests per day × days

Source

Per-million-token pricing as listed on AI providers' official pricing pages: cost = tokens × price per 1M tokens ÷ 1,000,000, with input and output priced separately. Prices change often, so enter the current figures from your provider.

Checked: 2026-11-22 · Results are for information only.

Common examples

Requests per dayMonthly cost
100$24.00
1,000$240.00
10,000$2,400.00
100,000$24,000.00

Select a row to load it into the calculator.

Good to know

  • Input tokens include everything sent with each request: system instructions, earlier chat turns and pasted documents. In a long conversation the input grows with every turn.
  • Many providers offer lower prices for batch jobs or for repeated input that is cached. Enter the discounted price per million tokens to see the effect.
  • Set a monthly spending limit or alert in your provider's account settings so a bug or a traffic spike cannot run up a large bill.

Questions and answers

Why do long answers raise the bill faster?

Most providers list a higher price for output tokens than for input tokens, so each extra token the model writes costs more than each extra token you send. The "share of the cost from output" line shows how much of your bill comes from answers.

Where do I find the price per million tokens?

On the official pricing page of the AI provider, listed for each model. Prices are usually given in US dollars per 1 million tokens, separately for input and output.

Related calculators