Chinese Word Counter — Words, Characters & Length Limits, Live
Free online Chinese word counter: hanzi + English word count, character count, and platform limits for X, SMS, Xiaohongshu and WeChat — runs in your browser.
About this tool
Free online Chinese word counter that gets mixed-script text right: paste anything and get the word-style count, character count, Chinese characters (hanzi), English words, sentences and paragraphs in one pass — plus a length-limit gauge that switches with your target platform: X posts 280, Chinese SMS 70, SEO description 160, Xiaohongshu 1000, WeChat article titles 64. Counting runs locally in your browser in real time — no sign-up, no upload, works offline.
Most word counters were built for English. Feed them Chinese and they collapse an entire paragraph into "1 word", because Chinese doesn't separate words with spaces. This tool counts the way Chinese editors and Microsoft Word's Chinese edition do: each hanzi counts as one, each run of Latin letters or digits counts as one — so a mixed sentence gets a number you can actually trust.
Words, characters, or word count — three different numbers
The same text produces different totals under different counting rules. Know which one you need:
| Metric | How it counts | Example: AI 时代 2026 |
|---|---|---|
| Words (word-style) | 1 hanzi = 1; each Latin/digit run = 1 | 4 (AI, 时, 代, 2026) |
| Characters (with spaces) | every character counted, spaces included | 10 |
| Characters (no spaces) | same, minus the spaces | 8 |
- Word-style count matches what Microsoft Word's Chinese edition reports as 「字数」 — the number thesis guidelines and Chinese editorial requirements refer to.
- Character count is what most platform input boxes enforce (X posts, SMS): every hanzi, letter, digit or space takes one slot.
- English words counts runs of letters (
don't= 1 word), tallied separately from hanzi so neither inflates the other.
All eight metrics are shown side by side, and the limit gauge re-bases itself on the right one when you switch scenes.
Platform length limits: X, SMS, SEO, Xiaohongshu, WeChat
Platforms don't agree on how to count — some use characters, some use words — and what happens when you overflow differs too:
| Platform / scene | Limit | Unit | If you exceed it |
|---|---|---|---|
| X (Twitter) post | 280 | characters | cannot post (longer for premium) |
| Chinese SMS (single) | 70 | characters | split into multiple billed messages |
| SEO title tag | ≈60 | characters | truncated in search results |
| SEO meta description | ≈160 | characters | truncated in search results |
| Xiaohongshu (Rednote) body | 1000 | words | cannot post |
| WeChat article title | 64 | words | cannot display in full |
Two details worth knowing:
- X counts characters, and one hanzi takes one character — so a Chinese post actually fits 280 hanzi, not "280 words" the way many counters assume.
- Xiaohongshu and WeChat count by 字 (word-style), close to Word's figure. A draft that fits Xiaohongshu can blow past an X post's limit instantly — that's exactly why the gauge switches with the scene.
How mixed Chinese-English text is counted
Mixed-script text is where naive counters fail. Take 「今天 launch 了 3 个 feature」:
- Word-style count = 7 (今, 天, launch, 了, 3, 个, feature)
- Character count = 24 (every character, spaces included)
- 4 hanzi, 2 English words, 1 digit run
The word-style total sums the hanzi and the English words — the same convention Word uses for mixed documents. Translators and localizers get language-split tallies for quoting and scheduling, not one lumped number.
Does the count match Microsoft Word?
The word-style count follows the same convention as Microsoft Word's Chinese 「字数」: hanzi counted one by one, each Latin or digit run counted as one. For everyday text the results match Word's word-count dialog.
Edge cases can differ slightly — Word has its own rules for a few full-width punctuation marks and symbols. For a thesis submission or journal requirement, verify once more in Word; for everyday writing and platform posts, this tool is accurate enough.
Reading time vs speaking time
The tool estimates both durations from median native-reader speeds:
- Reading: 400 hanzi/min for Chinese + 220 words/min for English
- Speaking: 240 hanzi/min + 130 words/min
Mixed text accumulates per language, then sums. Use speaking to time video scripts and podcast narration, reading for articles and posts — check before publishing so a 3-minute voiceover doesn't turn into 5.
Who uses this counter
- Bilingual content teams: one count for WeChat, one for X, no more surprises at the publish box
- Translators and localizers: language-split counts for quotes, invoices and scheduling
- Students and academics: abstract word limits under the Word convention
- SEO practitioners: titles and descriptions kept under 60/160 characters
- SMS and push copywriters: staying within the single-message 70-character window
To format and publish the text you just counted, pair it with the online Markdown editor; need a Traditional Chinese version too? Convert it in one pass with the Simplified-Traditional converter; see all tools in the toolbox.
FAQ
What is the difference between word count and character count?
WC_FAQ_A1
Which number should I read for a thesis or assignment?
WC_FAQ_A2
Do Xiaohongshu and WeChat count by words or characters?
WC_FAQ_A3
How is English text counted?
WC_FAQ_A4
Does the counter upload my text?
WC_FAQ_A5