Prepare your prompt
See roughly how many tokens your prompt uses, along with words, characters and reading time. Paste your prompt below.
Prepare your prompt
Language models such as ChatGPT and Claude do not read words — they read tokens. A token is a chunk of text, often a short word or part of a longer one. Both the context limit you can fill and the price you pay per request are measured in tokens, so knowing roughly how long a prompt is helps you stay inside the limit and keep costs predictable.
As a rule of thumb, English text averages about four characters per token, or roughly three tokens for every four words. This tool blends both rules to give a practical estimate that updates live as you type. Common words usually cost one token, while rare words, names, code and other languages get split into several.
Every model family uses its own tokenizer, so the exact number differs between ChatGPT, Claude, Gemini and open-source models. Treat the figure here as a close approximation for planning — ideal for checking whether a long prompt will fit, comparing two drafts, or budgeting a batch job. For billing-exact numbers, use the official tokenizer of the model you are calling.
Pair this with Markdown to plain text to strip formatting symbols that quietly eat tokens, or the readability checker to keep your writing clear.