MDUtil

Markdown Word Counter

Paste Markdown and count the words a reader actually sees, without #, **, link addresses or code. Nothing is uploaded.

Markdown

659 characters23 lines

Statistics

Words
80
Characters
451
Reading time
Under 1 min
Speaking time
Under 1 min
Characters without spaces
367
Sentences
11
Paragraphs
6

Structure

Headings
2
Links
1
Images
0
Tables
1
Code blocks
1
List items
3
Tasks done
1 / 1
Footnotes
1

How to count words in Markdown

  1. Paste text

    Paste Markdown, an AI answer or any text, or open a .md file.

  2. Choose what counts

    Switch on Count code blocks if the code inside them should count as words too.

  3. Read the numbers

    Words, characters, reading and speaking time and the document's structure update as you type.

What it handles

Only what readers see
Syntax characters, link addresses and image file names are left out of every count.
Chinese and Japanese counted like Word
Each Chinese or Japanese character counts as one word, as in Microsoft Word, and other text is split by the browser's word rules.
Reading and speaking time
Reading assumes 238 words or about 300 Chinese and Japanese characters a minute; speaking, 150 words or 220 characters.
Characters with and without spaces
Both counts are shown, since essay and form limits ask for one or the other.
Document structure
Headings, paragraphs, links, images, tables, code blocks, list items, tasks and footnotes are counted as well.
Code on request
Code blocks are left out of the word count unless you switch them on.

Count words in a Markdown file from the command line

Running wc -w on a .md file counts every #, ** and link address as a word. Convert the file to plain text with Pandoc first:

pandoc -t plain README.md | wc -w

wc splits words at spaces, so this works for English and similar languages but not for Chinese or Japanese, which this page counts character by character.

Questions and answers

How are words counted in Markdown?

The Markdown is turned into the text a reader sees first, so #, **, link addresses and image file names are not counted. The text is then split into words, and each Chinese or Japanese character counts as one word.

Does the count match Microsoft Word?

For most texts it does; Word also counts each Chinese and Japanese character as a word. Small differences come from hyphenated words and numbers with punctuation, which programs split in different ways.

How is the reading time worked out?

Words are divided by 238 a minute, the average silent reading speed measured for adults reading English, and Chinese and Japanese characters by about 300 a minute.

Are code blocks counted?

Not by default, since code is not read like prose. Switch on Count code blocks to include them.

Is my text uploaded anywhere?

No. The counting happens in your browser, so the text never leaves your device.