AI Token & Cost Estimator
Convert prompt and response word counts into estimated tokens and dollars, using separate input and output rates, for a single request or a whole batch.
Estimated tokens & cost
Tokens are estimated at about 0.75 words per token. Actual tokenisation and prices depend on the model and provider — check current rates before budgeting.
From words to tokens to cost
Language-model pricing is quoted per million tokens, split between the prompt you send and the text the model writes back. Because you usually think in words, not tokens, the first step is a conversion: roughly 0.75 words per token for English. From there the cost is simple arithmetic against the input and output rates.
This estimator is for budgeting and comparison — deciding whether a long prompt is worth trimming, or how a task scales across thousands of requests. It is not a bill: exact tokenisation and current prices come from your provider, so confirm them before you rely on a figure.
Frequently Asked Questions
What is a token?
A token is the unit a language model reads and writes — often a word piece rather than a whole word. For English text, a rough rule of thumb is about 0.75 words per token, so 1,000 words is roughly 1,300 tokens. This estimator uses that ratio to turn word counts into tokens.
Why are input and output priced separately?
Most providers charge one rate for the tokens you send (the prompt, or input) and a higher rate for the tokens the model generates (the completion, or output). Entering both rates gives a more realistic estimate than a single blended price.
How exact is the cost?
It is a planning estimate. Real token counts depend on the exact text and the model's tokeniser, and prices change and vary by provider and model tier. Always check current, official pricing before committing to a budget.
How do I estimate cost at scale?
Set the input and output word counts for a single typical request, then enter how many times it will run. The estimator multiplies the per-request cost by the number of requests so you can see the total for a batch or a busy day.
Estimates only. Token counts and prices depend on the model and provider; verify current rates before budgeting.