The text statistics tool counts four things about the text you paste: characters including spaces, characters excluding spaces, words and lines. It runs in one pass on this site’s own server and returns the numbers as JSON, so you can copy them into a document, a spreadsheet or a content brief. No AI model is involved and no credits are deducted, which makes this a free example of how a tool page works here — and a genuinely useful check when you write captions, alt text or prompts for the photo editor.
What each number means
Characters with spaces counts every Unicode code point in the text: letters, digits, punctuation, spaces, tabs and line breaks. An accented letter such as é is one character, a simple emoji such as 🙂 is one character, and a combined emoji built with zero-width joiners, such as a family emoji, counts as five or more characters because it is several code points joined together.
Characters without spaces removes everything that Unicode treats as whitespace, including ordinary spaces, tabs, line breaks and non-breaking spaces. Publishers and translation agencies sometimes quote a per-character rate that ignores spaces, so this is the number to compare against their limit or invoice.
Words splits the text on whitespace and discards empty pieces. A hyphenated term such as well-known counts as one word, a contraction such as don’t counts as one word, and a number such as 2026 counts as one word. Chinese, Japanese and Thai do not put spaces between words, so a whole sentence in those languages is counted as a single word; use the character counts instead when you work with those scripts.
Lines splits the text on Windows, old Mac and Unix line endings. A trailing newline creates one empty final line, and blank lines between paragraphs are counted as lines too. If the number looks one higher than you expect, remove the newline at the very end and run the tool again.
How to use the tool
Paste or type your text into the input box. The box accepts up to 50,000 characters per run, which is roughly a 15-page document or a long feature article. Press Run tool and the four numbers appear in the result panel as JSON. Use Copy to put the result on your clipboard, or Download to save it as a JSON file for a spreadsheet or a report.
The tool needs a signed-in account, like every tool on this site, but it costs zero credits, so running it never reduces your balance. Credits stay available for image edits on the photo editor. If you only need a quick check, paste a representative paragraph instead of the whole file; the proportions stay the same and the run is faster.
Where these numbers matter
Most publishing limits are expressed in characters, while editorial briefs are usually expressed in words. Knowing both numbers before you write saves a round of editing. These are the limits people ask about most often:
- X (Twitter): 280 characters per post. X counts most Chinese, Japanese and Korean characters as two, so check the composer as well as a plain character count.
- Instagram: 2,200 characters per caption, with only the first 125 or so shown before the More link.
- LinkedIn: 3,000 characters per post; the feed truncates after roughly 200 characters.
- YouTube: 100 characters for a title and 5,000 characters for a description.
- Google search results: a meta description is usually cut after about 155 to 160 characters on desktop and closer to 120 on mobile.
- Image alt text: write a short sentence of roughly 125 characters so screen readers finish it in one pass, and describe what the picture shows rather than repeating the file name.
- SMS: 160 characters for the GSM alphabet, but only 70 for a message that contains Unicode such as emoji or Chinese text.
- Journal and conference abstracts: commonly 150 to 300 words, depending on the venue.
- Translation quotes: usually priced per source word for European languages and per character for Chinese, Japanese and Korean.
- Reading time: adults read silently at roughly 200 to 250 words per minute, so 225 words per minute is a reasonable planning figure.
Counting text before an AI photo edit
The photo editor on this site reads a written instruction, so the text you write is the product input rather than decoration. A prompt that runs to several hundred words usually produces a muddier result than one that names the subject, the change and the style in two or three sentences. Counting words first is a cheap way to notice that you have stacked five competing ideas into one instruction; split them into separate edits instead.
The same numbers help after the edit. When you publish an image on a blog or a product page, the alt text and the caption are the parts search engines and screen readers actually read. Keep the alt text to one factual sentence, keep the caption inside the platform limits listed above, and keep the file name short and descriptive. None of that needs an AI model, only a count and a careful sentence.
Worked examples
A post that fills the 280-character limit on X is usually about 40 to 55 English words, depending on how many short words and spaces it contains. A 500-word article is roughly 2,800 to 3,200 characters with spaces. A 1,000-word blog post usually lands between 5,500 and 6,500 characters and takes four to five minutes to read at 225 words per minute. A 150-word abstract is about 900 to 1,000 characters, which is why it fits a single page even with a title and author list.
These ranges are guides, not rules. Sentence length, punctuation and the number of technical terms all move the ratio between words and characters. When a hard limit matters, paste the exact final text and trust the count rather than the estimate.
Frequently asked questions
Does the tool count an emoji as one character?
A simple emoji is one Unicode code point and counts as one character. Emoji that combine several code points, such as skin-tone variants or family emoji joined with zero-width joiners, count as several characters. The number matches the length of the text when it is read as Unicode code points, which is the same way most programming languages and databases measure it.
Does it work for Chinese, Japanese and Korean text?
Yes, but read the character counts rather than the word count. Those languages do not separate words with spaces, so the word counter treats a whole run of text as one word. The character counts, with and without spaces, are accurate and are the numbers publishers use for CJK copy.
Is my text stored or sent to an AI provider?
The text is processed in memory by this site’s own server code and is not saved to a database or sent to an AI provider. The tool is a local statistics example, not a generation feature. As with any online form, avoid pasting passwords, private keys or other secrets you would not put in a browser.
Why is the word count different from Microsoft Word or Google Docs?
Word processors use their own rules for hyphenated compounds, numbers, URLs and CJK text. This tool uses a simple whitespace split, which is predictable and easy to reproduce in a script. Small differences of one or two words are normal, especially around dashes and abbreviations.
Does it cost credits or require a subscription?
No. The text statistics tool is free and costs zero credits on every run. You do need to sign in, because all tools share the same account and rate-limit layer, but running it does not change your credit balance and does not require a paid plan. Credits are only used when you run an image edit.
How much text can I check at once?
Each run accepts up to 50,000 characters, which covers a long article, a report or a short story. For a book-length manuscript, split it into chapters and add the numbers together; the line count will include the breaks between chapters, so count lines per chapter if you need an exact figure.
Can I use the result in my own spreadsheet?
Yes. Use Download to save the JSON result and import it into a spreadsheet, or copy the values manually. The four fields are characters, words, lines and nonWhitespace, and they are stable, so a formula or script that reads them will keep working.