LLM API Cost Calculator (any model, your own token prices, any currency)

How it works

Cost per request = (uncached input tokens × input price + cached input tokens × cached price + output tokens × output price) ÷ 1,000,000. The discount reduces the cost of the chosen share of requests. Day = request cost × requests per day; month = day × days; year = month × 12. Use this page for any model or reseller, in any currency; the provider pages already contain official list prices.

⚠ Estimates for information only — not financial advice.

FAQ

Where do I find the token counts?

Every API response reports input and output tokens. Average them over a day of real traffic for the most reliable estimate.

My provider bills in another currency — does that matter?

No. Enter the prices in the currency you are billed in and pick that currency at the top; the arithmetic is the same.

Checked 10 Nov 2026. Per-token billing as described on the official pricing pages of OpenAI, Anthropic, Google, Mistral, DeepSeek, xAI and Meta (prices per 1 million tokens, separate input, cached-input and output rates).