FREE BROWSER TOOLAI workflow
LLM API Cost Calculator — Input and Output Token Budget
Estimate a batch of language-model requests before running it. Enter your own input and output prices in US dollars per million tokens; the preset rates are an arithmetic example, not a quote for any model.
Calculations run in your browser. No account or API key is needed.
Your result
- Input cost per request
- 0.002 USD
- Output cost per request
- 0.002 USD
- Total per request
- 0.004 USD
- Total for all requests
- 4 USD
Input cost per request: 0.002 USD Output cost per request: 0.002 USD Total per request: 0.004 USD Total for all requests: 4 USD Requests: 1,000 Estimate using your rates; additional provider charges are excluded.
How to use this tool
Enter typical input and output token counts for one request.
Replace both sample rates with the prices for your provider, model and billing tier.
Set the request count and compare the input, output and combined estimates.
How the calculation works
Each token count is multiplied by its corresponding rate and divided by 1,000,000. The two costs are added, then multiplied by the request count. With 2,000 input tokens at $1 and 500 output tokens at $4 per million, one request costs $0.004 and 1,000 requests cost $4.
Use comparable measurements
Use token counts from representative API responses rather than word counts. Compare a short-answer scenario with a longer-answer scenario to see which part drives the budget. This simple calculation excludes cache pricing, batch discounts, tools, images billed separately, taxes and retries; account for those separately when applicable.
Common questions
Are the prices updated automatically?
No. Enter current rates from your provider. Different endpoints and billing tiers can have different charges.
Does the estimate include reasoning tokens?
Only if you include the billable reasoning tokens in the output count. Check your provider’s usage report.