Word & Character Counter

Text is split at spaces and line breaks, so "well-known", "don't", "3.5" and a web address each count as one word, as in a word processor. Anything with no letters or digits, such as a dash on its own or an emoji, is not a word. Chinese, Japanese and Thai are written without spaces, so those parts are split into words with your browser's built-in word segmenter (Intl.Segmenter), which uses a dictionary: 我喜欢学习中文 counts as four words (我 / 喜欢 / 学习 / 中文), not one and not seven. Japanese tends to split into smaller pieces, particles included. Counts for those languages can differ a little from other tools, because there is no single agreed way to split them.

Characters are counted the way you see them on screen (grapheme clusters, via Intl.Segmenter), so an emoji, a flag, a family emoji built from several code points, or an accented letter typed as two code points each count as one character. Many simple counters use the raw string length instead, which counts 👍 as 2 and 👨‍👩‍👧 as 8. Some platforms do count that way, which is why the X and SMS bars use their own rules.

Sentences are found with the browser's sentence segmenter, which ends a sentence at . ! ? (and their Chinese and Japanese forms) or a line break. Abbreviations such as "Dr. Smith" can be counted as a sentence break, so treat the number as close rather than exact. Paragraphs are blocks of text separated by a blank line; lines counts every line, blank ones included.

Words divided by 238 words per minute: the average silent reading rate for English non-fiction in Marc Brysbaert's 2019 meta-analysis of 190 studies (Journal of Memory and Language, "How many words do we read per minute?"). Speaking time uses 150 words per minute, a typical pace for a talk or presentation rather than a measured figure; your own pace may be quite different.

Each bar links to the official page it comes from (checked October 2026): X counts most Latin characters as 1, CJK characters and emoji as 2 and every link as 23, up to 280. LinkedIn posts allow 3,000 characters and Instagram captions 2,200 (plus at most 30 hashtags). The meta description bar is a guideline only: Google says there is no length limit and it trims snippets to fit the screen.

A standard SMS carries 140 bytes. Text that only uses the GSM-7 alphabet (basic Latin letters, digits, common punctuation and a few accented letters) fits 160 characters; a single character outside it, such as a curly quote ’, an emoji or a Cyrillic letter, makes the phone send the whole message as UCS-2, which fits 70. Longer messages are split into parts of 153 (GSM-7) or 67 (UCS-2). The bar shows which characters caused the switch, so you can replace them.

No. Everything is counted in your browser and nothing you type is sent anywhere. The text is only kept if you tick "Remember this text on this device" (off by default): it is then stored in this browser's local storage until you untick it or clear it.

Flesch Reading Ease is 206.835 - 1.015 x (words per sentence) - 84.6 x (syllables per word), from Rudolf Flesch's 1948 paper; higher is easier, and Flesch labelled 60-69 Standard, 50-59 Fairly difficult and under 30 Very difficult. Flesch-Kincaid Grade Level is 0.39 x (words per sentence) + 11.8 x (syllables per word) - 15.59, from Kincaid and colleagues' 1975 report for the US Navy, and reads as a US school grade. Syllables are estimated from spelling, so a word like "requirements" can be off by one. Both formulas were made for English: the panel hides them when most of the text isn't in Latin script, and they mean little for French, Spanish or German either.

Not as a file - this opens .txt and .md files only. Open the document, select all the text, copy it and paste it in. For a PDF you can't copy from, our PDF to Word tool can extract the text first.