Free tool · Updated 2026

What is your LLM API actually costing you?

Enter your usage, compare every major model side by side, and see how much you'd save by routing cheap work to a cheaper model. No signup.

Your monthly usage

API calls per month
Prompt + context sent
Tokens the model returns
potential monthly saving vs running everything on your premium model.
All-premium:  →  Routed:  (/mo saved)

Routing estimate

Share of traffic that truly needs the premium model20%

Cost by model — cheapest first

Model $ / 1M in $ / 1M out Monthly cost vs premium
Prices are editable defaults. Claude figures reflect published per-token pricing; others are representative estimates. Click any price to edit it, then confirm against each provider's own pricing page before relying on the numbers.
PRO WORKBOOK

Get the full cost model as a spreadsheet

Everything here plus 12-month projections with growth, saved routing scenarios, per-model comparison, and a self-host break-even calculator. Editable Excel workbook — yours to keep.

Get the Pro workbook — $19

Save this automatically

A routing layer sends each request to the cheapest model that can handle it — capturing the saving above without you picking a model per call.

See routing options →