Character Counter Guide: Exact Characters, Platform Limits and the SMS Trap – AI Humanizer

What a character counter is, and why “a character” is not one simple thing A character counter measures text in its smallest visible units. That sounds obvious until you ask the only question that matters: what exactly is one character? A reader sees one thing in the letter a, one thing in an accented é, and one thing in a smiling emoji. A computer sees three different storage stories. The letter a is one Unicode code point, one UTF-16 unit and one UTF-8 byte. An é written as a base letter plus a combining accent is one visible character but two code points. A family emoji joined with invisible joiner characters can be one picture made of seven code points, eight UTF-16 units and twenty-five bytes. Every counter on the internet quietly picks one of these definitions, and most never tell you which. That quiet choice is why two honest tools can disagree about the same text, and why a post that “fits” in one editor gets rejected by a platform. When the limit is loose — an essay, a caption with room to spare — nobody cares. When the limit is tight — a 280-character post, a 150-character bio, a 30-character ad headline, an SMS that bills by the segment — the definition is the whole game. This guide explains how this counter handles that honestly: it shows the main counts side by side, names the method behind each one, and checks your text against the limits people actually write for. What this character counter does, and what it does not do This tool counts in your browser, live as you type, paste or open a file. The headline number is Characters: grapheme clusters, the units a reader actually sees, counted with your browser’s Unicode segmentation. Alongside it you get characters without spaces, Unicode code points, UTF-16 code units (the length JavaScript and many form fields use), UTF-8 bytes (the storage size that file, database and API limits measure), a word and line count for orientation, and a count of the part you have selected. Underneath sit the platform presets: pick one and the tool shows how much of the limit you have used, how much is left, and turns the bar red when you cross it. There is a custom limit for everything else, and an SMS panel that tells you whether your message still fits the old 160-character world or has already slipped into 70-character Unicode segments. Now the honest list of what it does not do. It does not send your text anywhere: no upload, no account, no history waiting for you tomorrow. It does not promise that a platform will accept your text, because platforms weigh characters in their own way and change their limits without asking anyone. It does not fix your text, shorten it, or judge it. And it is deliberately not the Word Counter: that tool answers how much you wrote — words, sentences, reading time, repeated words. This one answers whether it fits, and how it is being counted. The two pages link to each other so you can jump to the question you actually have. How to count characters step by step Start by putting your text in the box. Type it, paste it from the app you are drafting in, or open a plain .txt, .md or .csv file from your device; the file is read on your device, never uploaded. Watch the eight counters settle — on a paragraph of ordinary prose most of them will agree, and the moment they stop agreeing you have emoji, accented letters, joined scripts or symbols in play, which is exactly the moment this tool earns its keep. If only part of the text matters — the headline, the first line, one paragraph — select just that part and read the selection strip instead of re-counting the whole document in your head. Then choose the limit you are writing against. The preset row covers the everyday ones: an X post, Instagram caption and bio, TikTok caption, LinkedIn and Facebook posts, YouTube title and description, meta title and description, and Google Ads lines. Anything else — an exam form, a marketplace field, a database column — goes in the custom limit box. The bar and the remaining count answer the only question that matters: does it fit, and by how much? If it does not fit, trim against the live number instead of guessing, then copy the text or download it as a .txt. One habit worth building: check the SMS panel before sending any long message that mixes scripts or emoji, because that is where the quiet 160-to-70 shrink ambushes people. The four counts explained with honest examples Take the word “café”. Written the common way it is four graphemes, four code points, four UTF-16 units and five UTF-8 bytes, because the é needs two bytes in UTF-8. Written the other way — e plus a combining accent — it still looks identical and still reads as four graphemes, but now it is five code points, five UTF-16 units and six bytes. Nothing about the text changed for the reader; everything changed for storage. That is the whole lesson of this page in one word: the visible character and the stored size are different facts, and different limits care about different facts. Emoji push the gap wider. A simple smiley is one grapheme, one code point, two UTF-16 units and four UTF-8 bytes. Add a skin tone and it becomes two code points glued together. A family emoji is several person emoji welded by zero-width joiners into a single picture — one grapheme, seven or more code points. A form that counts “characters” with the browser’s built-in length is counting UTF-16 units, so your one family picture already costs two per person inside it. This counter shows all four numbers precisely so none of this is a surprise: reader-facing text uses the grapheme count, technical limits use bytes, and anything built on JavaScript string length uses UTF-16. Platform limits, presets, and how to use them honestly The presets exist because the limits are real but badly scattered. An X post allows 280 characters. An Instagram caption allows 2,200 and a bio only 150. TikTok captions are commonly published at 2,200. LinkedIn posts run to 3,000, Facebook posts to a rarely-reached 63,206. A YouTube title stops at 100 characters and a description at 5,000. Meta titles and descriptions are a different species: Google truncates by display width, not a hard count, so the roughly 60 and 160 figures are planning recommendations and the tool labels them that way. Google Ads is strict again: 30 characters for a headline, 90 for a description. Every figure here is the commonly published one, shown with the same warning the tool shows: platforms change these numbers when they like. Two habits keep the presets honest. First, remember that the app itself is the final referee. X counts some scripts and emoji double, links get wrapped and counted at a fixed length, and line breaks are not free everywhere. If the platform’s own counter disagrees with this page at the moment you post, the platform wins — the presets are for planning your draft, not for arguing with a compose box. Second, do not write to the last character. Limits change, counters differ by a little, and a caption composed at exactly 2,200 is a caption one quiet platform update away from being cut. Leave a margin and your text survives the next change. The SMS trap and the emoji disagreement SMS is where character counting stops being academic. The classic text message uses the GSM-7 alphabet: the letters, digits and common punctuation of Western European languages, 160 characters in one segment. The trap is that this alphabet is small. One character outside it — anything in Urdu or Arabic script, most emoji, many symbols — and the entire message is re-encoded as Unicode, where a segment holds only 70 characters. A 150-character message written in Latin letters is one segment; the same message with a single Arabic word inside it is three segments. Worse, longer messages glue together from 153- or 67-character pieces, so the billable segments multiply faster than the headline numbers suggest. The SMS panel on this page detects which world your text lives in and shows units, segment size and segment count before you send. Emoji cause the quieter disagreement. A platform counting grapheme-style sees one family emoji; a form field counting UTF-16 units sees eight; a database counting bytes sees twenty-five. None of them is lying — they are answering different questions. The disagreement only hurts when you plan with one number and the gatekeeper uses another. That is why this tool refuses to show a single “true” count: it shows the reader count for social limits, the byte count for storage limits, and names the method on each tile so you can match the number to the limit that will actually judge your text. Privacy, limits of this tool, and the questions people ask Privacy is short because there is little to say: everything is counted in this browser tab. Your text is not uploaded, not stored on a server, not shared, and closing the tab forgets it — test that by loading the page, switching off Wi-Fi and mobile data, and counting away. What the tool cannot do follows from the same design: it cannot remember your last draft, sync across devices, or warn you tomorrow that a platform changed its limits. The preset figures are the commonly published limits at the time of writing, labelled as changeable, and the grapheme count depends on your browser’s Unicode support; very old browsers fall back to code points, and the page says so when that happens. The questions people ask are predictable. Is it free? Yes, with no account. Do spaces count? In the main count yes — they are characters a reader’s eye crosses — and the no-spaces count sits beside it for the forms that ask the other way. Which number should I trust? The one your limit uses: graphemes for reader-facing platform limits, bytes for file and API limits, UTF-16 for most form fields. When should I use the Word Counter instead? When the question is how much you wrote rather than whether it fits — words, sentences, paragraphs, reading time. Used for its own job, a character counter is a small honest machine: it does not make your text better, it makes its size impossible to misunderstand.

Character Counter Guide: Exact Characters, Platform Limits and the SMS Trap – AI Humanizer

What a character really is (grapheme, code point, UTF-16 unit, byte), how to count text against X, Instagram, LinkedIn, meta and ad limits, why SMS drops from 160 to 70 characters, and how this counter keeps every method honest.