Placeholder text with an exact number of paragraphs, sentences, words or characters
The Lorem Ipsum Generator makes filler text for the empty spots in screen and print mockups. Pick a unit (paragraphs, sentences, words or characters), enter an amount, and the proof shows text that matches it exactly. Choose 3 paragraphs and you get exactly 3 paragraphs separated by blank lines. Choose 300 characters and you get exactly 300 characters, spaces included, ending with a period.
There are three scripts. The default, Lorem ipsum, is the classic placeholder text derived from a Latin passage by Cicero. The 한글 option strings together common Korean syllables into meaningless words, each followed by a particle that fits its final sound. The 日本語 option mixes sentences from a novel by Miyazawa Kenji that is published by Aozora Bunko. If you are designing a Korean or Japanese screen, dummy text in the real script shows line breaks and character widths more faithfully than Latin letters do.
The ruler under the proof shows how many characters fit on one line at the typeface, size and column width you picked. When the result looks right, copy it as plain text or as HTML paragraphs, or copy a link so someone else sees the same text.
Why placeholder text is needed and when to use it
Layout work has to start before the final copy exists. To size a box and set its margins when you do not know whether the text under a heading will run two lines or ten, you need fake text of about the same length. Real sentences pull reviewers toward judging the wording instead of the layout, so meaningless text fills the shape and nothing more. This tool is most useful in these situations.
- Web and app mockups: put in the expected amount (2 sentences for a card description, 5 paragraphs for an article) and check card heights and scroll length.
- Testing fields with a character limit: if an input or a forum post caps text at 300 or 1,000 characters, generate exactly that many characters and check overflow and ellipsis handling. If the cap is in bytes, look at the UTF-8 bytes in the statistics.
- Comparing typefaces, sizes and columns: switch the same text between sans and serif, 14 to 20px, and 1 to 3 columns, and see whether the line length stays in a comfortable range.
- Korean and Japanese editions: Latin letters are narrower than Korean or Japanese ones and are spaced differently, so the same number of characters covers a different area. Pick 한글 or 日本語 to fill the layout at a density close to the real thing.
- Sample data for development: paste HTML paragraphs into a template that needs several paragraphs, and share a link that reproduces the same sample.
How to use it
- Pick a unit. Paragraphs, sentences, words or characters. Changing the unit resets the amount to that unit's default (3 paragraphs, 5 sentences, 100 words, 300 characters).
- Enter an amount. Type it in the number field, or use the − and + buttons or the common amounts next to it. The proof updates as you change the number.
- Pick a script. With Lorem ipsum you can choose whether the text starts with 'Lorem ipsum dolor sit amet, consectetur adipiscing elit.' Turn it off and the text is random from the first sentence.
- Check the proof. Change the sans or serif face, the font size and the number of columns below it, and read the ruler's 'About n characters per line'. These controls never change the text itself.
- Copy it. Use Copy text for documents and Copy HTML for web markup. For different text, press Generate new and the same settings give you a fresh text.
What each unit matches exactly
| Unit | What is matched exactly | Range | Default | Good for |
|---|---|---|---|---|
| Paragraphs | Paragraphs separated by a blank line | 1–50 | 3 | Article and blog body text, scroll length |
| Sentences | Sentences ending in a period (in a single paragraph) | 1–200 | 5 | Card descriptions, helper text, empty states |
| Words | Words separated by spaces (Korean: space-separated chunks) | 1–2,000 | 100 | Copy with a set word count, summaries |
| Characters | Characters without line breaks (spaces included) | 1–10,000 | 300 | Fields with a character limit, meta descriptions |
Only the unit you choose is exact; everything else depends on the text. Choose 100 words and you get exactly 100 words, but the number of sentences varies each time. In the character unit, the last sentence is cut to fit the space that is left and closed with a period, so its final word may be replaced by another word or become a short fragment. Japanese has no spaces between words, so the word unit is not used there, and you cannot select Words while Japanese is chosen.
How long is one paragraph or sentence: averages by script
Knowing roughly how many characters a paragraph or sentence gives you makes it easier to pick an amount. The table below shows averages over 6,000 texts generated with this tool for each script (with the opening sentence turned off).
| Script | Sentences per paragraph | Words per paragraph | Characters per paragraph | Characters per sentence | UTF-8 bytes per paragraph |
|---|---|---|---|---|---|
| Lorem ipsum | about 5.5 | about 55 | about 400 | about 72 | about 400 |
| 한글 | about 5.5 | about 49 | about 200 | about 36 | about 500 |
| 日本語 | about 6 | — | about 175 | about 29 | about 520 |
A Korean paragraph ends up with roughly half the characters of a Lorem ipsum one, yet it takes more bytes. UTF-8 stores basic Latin letters in 1 byte, and Korean syllables, kana and kanji in 3 bytes. If a system limits storage or transfer in bytes, test with the UTF-8 bytes in the statistics rather than the character count. By words, 100 words of Lorem ipsum average about 730 characters, and 100 Korean word units average about 410.
Reading the ruler: line length
If a line is too long, the eye struggles to find the start of the next one; if it is too short, lines break so often that the reading flow is interrupted. In The Elements of Typographic Style, the typographer Robert Bringhurst writes that 45 to 75 characters is widely regarded as a satisfactory line length for a single-column page set in a serif face at text size, and he names 66 as ideal. Success Criterion 1.4.8 (level AAA) of the web accessibility guidelines WCAG 2.2 requires that a block of text can be set no wider than 80 characters, or 40 for Chinese, Japanese and Korean.
The ruler under the proof measures the average character width of this text in the current typeface and size, draws a tick every 5 characters and a number every 10. The small triangle at the end marks the right edge of the column, and the character count at that spot is shown as 'About n characters per line'. For Lorem ipsum, the 45–75 range is shaded gray and anything beyond 80 is orange; for Korean and Japanese, anything beyond 40 is orange. With several columns, each column gets its own ruler.
| Verdict | Criterion | Meaning |
|---|---|---|
| Short measure | Under 45 characters (Latin) | Lines wrap often. Common in narrow cards and multi-column layouts on phones. |
| Comfortable measure | 45–75 characters (Latin) | The range Bringhurst calls satisfactory. |
| A bit long | 76–80 characters (Latin) | Within the WCAG limit, but over 75. |
| Over the WCAG limit of 80 | 81 or more characters (Latin) | Common in a single column on a wide screen. Split into columns or narrow the text block. |
| Within the WCAG limit (40 for CJK) | 40 or fewer (Korean, Japanese) | Inside the WCAG 1.4.8 limit for Chinese, Japanese and Korean text. |
| Over the WCAG limit (40 for CJK) | 41 or more (Korean, Japanese) | Common when Korean or Japanese body text is set in one wide column. |
These values are estimates from an average width. Real lines break at word boundaries, so they usually end a few characters short; text with many narrow letters such as i and l fits more, and text with many wide ones such as m and w fits fewer. Measured at the same width (sans-serif, 16px), a line holds about two thirds as many Korean characters as Lorem ipsum characters, and about half as many Japanese ones.
Why Korean and Japanese dummy text exist
Latin letters vary in width and are mostly narrow, while Korean syllables and kana or kanji each fill a nearly square cell, so far fewer of them fit in the same width. That is why a 300-character box sized with Lorem ipsum gets many more lines once 300 real Korean characters go in. If you are designing the Korean or Japanese edition of a screen, this matters. Line breaking differs too. With the browser default (CSS word-break: normal), a Korean line may break between any two syllables, and with keep-all it breaks only at spaces. CSS Text Module Level 3 describes keep-all as another approach commonly used for Korean text that has spaces.
The Korean dummy text strings together common syllables, 317 in all, into mostly two-syllable words at random, then adds a particle chosen by the final sound of the previous syllable: the forms eun, i, eul, gwa and euro after a syllable that ends in a consonant, and neun, ga, reul, wa and ro after one that does not (after a final l sound it uses ro, not euro). Sentences run 6 to 12 words and end with the plain verb ending -da. It has no meaning, but the particles and endings give it a rhythm close to real Korean, and combinations that happen to read like an insult or slur are filtered out and regenerated.
The Japanese dummy text mixes 91 narrative sentences, with dialogue and notes removed, from Miyazawa Kenji's novel 「ポラーノの広場」 (he died in 1933), which Aozora Bunko publishes. The Aozora Bunko handling guidelines state that files of works whose copyright has expired may be copied and used freely, and this tool credits the source and the typist and proofreader in the sources below. These are real sentences from a novel, so keep in mind that a reviewer who reads Japanese will understand them.
Where Lorem ipsum comes from
… neque porro quisquam est, qui dolorem ipsum, quia dolor sit, amet, consectetur, adipisci velit, sed quia non numquam eius modi tempora incidunt, ut labore et dolore magnam aliquam quaerat voluptatem.
Lorem ipsum dolor sit amet, consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.
Lorem ipsum starts with De finibus bonorum et malorum, a book on ethics that the Roman statesman and philosopher Cicero wrote in 45 BC. Section 32 of book 1 contains the passage saying that no one loves or pursues pain itself (dolorem ipsum) because it is pain. The first two letters of dolorem fell away, and lorem ipsum became the opening of the placeholder text. That is why lorem is not a word in any Latin dictionary.
The first person to trace this link was Richard McClintock of Hampden-Sydney College in Virginia. According to The Straight Dope, he looked up the unfamiliar word consectetur in a Latin dictionary, went through its uses in classical literature and recognized this passage in Cicero. Wikipedia sums up that he first reported it in a 1994 letter to the editor of the design magazine Before & After.
It is often said that the text has been used since a printer in the 1500s scrambled type to make a specimen book, but that claim spread from a guess in the same letter, and no printed example from that period has been confirmed. Wikipedia notes that McClintock later admitted it was an unsupported guess. The documented use is the Letraset transfer sheets of the late 1960s, and it spread widely when desktop publishing programs such as Aldus PageMaker included the text as a sample.
Common mistakes
- Trying to translate Lorem ipsum: the words were cut up and reordered, so it does not work as a sentence with a meaning. If you want to know the original meaning, see the Cicero text above.
- Testing a character limit by word count: 100 words of Lorem ipsum average about 730 characters and 100 Korean word units about 410, so the same 100 gives very different lengths. For fields with a character limit, test in the character unit.
- Judging a Korean screen to the end with Latin dummy text: a different number of characters fits in the same width, so real copy adds lines. Do the final check with Korean dummy text or the real copy.
- Letting placeholder text go live: before publishing, search the site for 'Lorem' or 'ipsum' to make sure no sentences are left. If it stays on a public page, it can also appear as the description in search results.
- Running user tests with dummy text: Nielsen Norman Group points out that filler such as Lorem ipsum is convenient early in a design but ends up causing problems when the real content is different in nature. Run tests in which people read the content and make a judgment with real copy.
- Pasting copied HTML into a plain-text editor: the <p> tags show up as literal text. For ordinary documents, Copy text is the right choice.
Limits
- Characters are counted as Unicode code points, with spaces included and line breaks left out. The text this tool makes contains no combining characters or emoji, so the count matches what you see on screen.
- Characters per line is an estimate from the average character width. Where a real line ends varies by a few characters with word length, typeface and browser.
- The proof uses the browser's default line breaking (word-break: normal). If your site uses word-level breaking (keep-all), Korean line ends become more ragged.
- The serif face uses a font installed on the device. On a device without one, another font is substituted and line lengths can change.
- The Japanese dummy text has only 91 sentences, so very long output repeats sentences. The Korean dummy text adds particles that fit the final sound, but its spacing and grammar are not those of real Korean.
- The text is made inside your browser and is not stored anywhere. A copied link holds only the settings and a seed number, and reloading returns to the settings carried in the link.
- If you choose 1 character, there is no room for a full stop, so you get a single letter without one.
As of 2026-10-22. The Cicero text, WCAG 2.2 Success Criterion 1.4.8, the Bringhurst quotation and the Aozora Bunko guidelines were checked against the original sources listed below.
Sources
- 01The Latin Library: Cicero, De Finibus Bonorum et Malorum IBook 1 [32] 'qui dolorem ipsum, quia dolor sit, amet, consectetur, adipisci velit…'
- 02The Straight Dope: What does the filler text "lorem ipsum" mean?A treatise on ethics from 45 BC, how McClintock traced the source through consectetur, Letraset sheets, Aldus PageMaker
- 03comp.fonts FAQ: Lorem ipsumQuotes Before & After vol. 4 no. 2 (McClintock's letter, mention of the 1500s)
- 04Wikipedia: Lorem ipsumThe 1994 letter, the 1500s date as an unsupported guess, Letraset transfer sheets of the late 1960s
- 05W3C: WCAG 2.2 Success Criterion 1.4.8 Visual Presentation (AAA)'Width is no more than 80 characters or glyphs (40 if CJK)'
- 06The Elements of Typographic Style Applied to the Web 2.1.2: Choose a comfortable measureBringhurst quoted: 45 to 75 characters is satisfactory, 66 is ideal
- 07W3C: CSS Text Module Level 3, word-breakkeep-all: another line breaking approach common for Korean text that uses spaces
- 08IETF RFC 3629: UTF-8U+0000 to 007F take 1 byte; U+0800 to FFFF (Korean syllables, kana, kanji) take 3 bytes
- 09青空文庫: 宮沢賢治「ポラーノの広場」底本 講談社文庫『銀河鉄道の夜・風の又三郎・ポラーノの広場 ほか三編』(天沢退二郎 編), 入力 白川由紀子, 校正 須藤
- 10青空文庫: 収録ファイルの取り扱い規準Works whose copyright has expired may be freely copied, redistributed and used
- 11Nielsen Norman Group: Fake Copy (2017)Filler text is handy early in design but causes problems when the real content differs in nature
Frequently asked questions
What does Lorem ipsum mean?
It is not a Latin sentence that makes sense. It was cut and altered from a passage in book 1, section 32 of Cicero's De finibus bonorum et malorum (45 BC), which mentions 'dolorem ipsum' (pain itself). Some words, such as lorem, which lost the first two letters of dolorem, are not in any Latin dictionary.
Does the character unit count spaces and line breaks?
It counts spaces but not line breaks. A result in the character unit is a single paragraph with no line breaks, and choosing 300 gives exactly 300 characters including spaces, ending with a period. The count without spaces and the UTF-8 bytes appear separately in the statistics.
How is the Korean dummy text made?
It strings together 317 common Korean syllables at random into words, then adds a matching particle: eun, i or eul after a syllable ending in a consonant, neun, ga or reul after one that does not. Sentences are 6 to 12 words long and end with the plain verb ending -da. It has no meaning, and combinations that happen to read like an insult are filtered out and regenerated.
Where does the Japanese dummy text come from?
It mixes 91 narrative sentences, with dialogue removed, from Miyazawa Kenji's novel 「ポラーノの広場」 (he died in 1933), published by Aozora Bunko. Japanese has no spaces between words, so it is matched by paragraphs, sentences or characters instead of words.
Do the same settings always give the same text?
Until you press Generate new, the same text stays, and raising the amount keeps the beginning and adds to the end. Opening a link made with Copy link shows the same settings and the same text. The address holds only the settings and a seed number, not the text itself.
How does the ruler work out 'about n characters per line'?
It measures the average character width of this text in the current typeface and size, then divides the column width by that value. Real lines break at word boundaries, so they can end a few characters short. The 45-75 band is Bringhurst's recommended range, and the 80 limit (40 for CJK) comes from WCAG 2.2 Success Criterion 1.4.8.
How is Copy HTML different from Copy text?
It wraps each paragraph in <p> and </p> and turns the &, < and > symbols into character references. Use it when pasting into web markup or a template. In a plain editor such as Notepad the tags appear as literal text, so Copy text suits ordinary documents.
Can I publish the generated text as it is?
It is best used only as temporary material for testing layouts. Before publishing, swap in real headings and sentences, and search the site for words like 'Lorem' and 'ipsum' to see if any remain. Tests where people read the content and make a judgment should also use real copy to be trustworthy.