Enter your monthly token usage — instantly see the cost per model, ranked cheapest first. All prices are editable and everything runs in your browser.
| # | Model | $/M in | $/M out | Est. monthly cost |
|---|
Prices are indicative, as of 2026-07 and vary by tier / caching / batch. Click any price cell to correct it, then the ranking recalculates. Always confirm on the provider's official pricing page. Nothing is uploaded — this is a client-side estimator.
API pricing is quoted per million tokens, split into input (your prompt) and output (the model's reply). Since output is usually 3–5× the input price, your real cost depends heavily on how much the model writes. Enter your monthly input/output volume above and this calculator ranks every model by total estimated spend.
A model that's 5× cheaper but needs 2× the tokens or more retries can cost more in practice. Use this as a first-pass estimate, then test the shortlist on your real workload.