How it works
What we count
- Words: sequences of letters/numbers separated by whitespace or punctuation (Unicode-aware enough for everyday English; other scripts may differ slightly by browser).
- Characters with spaces: every character in the box, including spaces and line breaks.
- Characters without spaces: the same count minus whitespace (spaces, tabs, newlines).
- Sentences: splits on common end punctuation (
.?!…) followed by a boundary; abbreviations can slightly over-count. - Reading time: based on about 225 words per minute for English prose. Adjust expectations for dense technical text.
Token estimate
We also show a rough token estimate (about 4 characters per token for English). It is not a model-accurate count. For exact o200k_base counts and prompt cost, use the token counter. To tidy AI output before counting, try the AI text cleaner.
FAQ
Is my text uploaded?
No. Counting runs entirely in your browser. You can confirm in the network panel: typing triggers no requests that send your text.
How is reading time calculated?
We divide the word count by 225 words per minute (a common estimate for adult English reading). It is a guide, not a guarantee.
Why show a token estimate here?
So you can see order-of-magnitude size before opening the token counter. For model-accurate tokens and cost, use the dedicated token counter.
Does this replace a character counter?
No separate character-counter page. This tool shows characters with and without spaces on the same URL.