What your prompt actually costs, across every major model
FAQ
How accurate is the token count?
It uses a character and word heuristic that lands within roughly 10% of real BPE tokenisers for English prose. Code and non-Latin scripts tokenise less efficiently, so treat it as an estimate.
Are the prices current?
Prices are published per million tokens and change often. Always confirm against the provider before budgeting anything important.
Does anything leave my browser?
No. There is no network call at all — you can use this offline.