Glitch Text Calculator - Zalgo & Cursed Unicode Maker

A glitch text calculator is an online utility that turns plain text into Zalgo or cursed Unicode by stacking combining diacritics above, below, or on both sides of each base character. You see the same look in memes, creepy forum posts, and dramatic chat messages, and the calculator lets you control how heavy the effect gets before you copy the result. The styled output is real Unicode, so it travels through plain text fields where CSS is unavailable. Most glitch text calculator users paste the result directly into a chat app or a username field.

• • Social media flair: Decorate a username, bio, or short comment with stacked Unicode so it stands out.
• • Meme and creative writing: Drop the styled text into horror fiction, surreal memes, or a screenshot caption.
• • Test Unicode handling: Check how an app, game, or form field handles long sequences of stacked marks.
• • Quick Zalgo practice: Compare direction and intensity settings to see how each one changes the look.

The calculator runs in the browser, so the styled text is ready as you type and the count of stacked marks updates at the same time. That is also why some apps strip the marks to keep their data clean, which is why the plain preview matters.

Zalgo text takes its name from a 2004 creepypasta by Dave Kelly whose central entity speaks in glitched, bleeding letters. The style became a meme shorthand for uncanny writing, and the calculator automates the trick of stacking marks until the text looks possessed.

To see the same base text rewritten in a different stylistic direction, the Lowercase to Uppercase Converter applies a related case-flipping transformation in your browser and counts the new characters as you type.

The calculator reads the input as Unicode, then for every base character it samples a small number of combining diacritics from the appropriate Unicode block and appends them in order. The result is one or more combined sequences where the marks stack on top of or below the original letter, digit, or punctuation.

• sourceText: The plain text typed or pasted into the form, processed as Unicode code points up to 240.
• intensity: Maximum combining marks sampled for each base character, drawn as a random integer in the 0 to intensity range.
• direction: Above-base pool (U+0300 to U+036F), below-base pool (U+0483 to U+0489 plus U+20D0 to U+20FF), or both combined.
• seed: Small integer that seeds a Mulberry32 pseudo-random generator so the same input and seed reproduce the same styled text.

The above-base pool is the Unicode Combining Diacritical Marks block; the below-base pool is the Cyrillic combining marks plus the Combining Marks for Symbols block. When direction is both, the calculator concatenates both pools before sampling.

For each base character the calculator draws a random integer between 0 and intensity, then draws that many extra random integers to pick which marks to append. A seeded random generator matters because it lets you re-create the exact same Zalgo text later by re-entering the same source, intensity, direction, and seed.

According to the Unicode Consortium, the Combining Diacritical Marks block (U+0300 to U+036F) and the Combining Marks for Symbols block (U+20D0 to U+20FF) define the characters that can be stacked above or below a base character to form Zalgo-style text.

According to Unicode Consortium, the Combining Diacritical Marks block (U+0300 to U+036F) and the Combining Marks for Symbols block (U+20D0 to U+20FF) define the characters that can be stacked above or below a base character to form Zalgo-style text.

If you want to see the code point and UTF-8 byte details for the same styled string, the Unicode Text Calculator provides the adjacent Unicode inspection workflow.

Glitch text sits on top of three Unicode ideas: the base character, the combining marks that follow it, and the receiving app that renders the combined sequence.

The base character is what shows up in a plain preview after every combining mark is stripped, so it is a useful sanity check that the source text was read correctly.

Stacking order matters because some apps limit the maximum number of stacked marks they will render. If the styled text appears cut off vertically, that limit is being hit, not a problem with the calculator itself.

Because Zalgo text often hides the underlying words, the Word Count Calculator is a useful follow-up to count the readable base characters after the marks are stripped.

Run the form, copy the styled text, paste it where you want it, and adjust the controls if the effect is too subtle or too heavy.

1. 1 Type or paste your text: Enter any text in the source field. The calculator keeps the first 240 Unicode code points and ignores the rest.
2. 2 Pick an intensity: Choose how many combining marks to sample per character. Start around 4 to 6 for a balanced look, raise for heavier Zalgo, drop to 0 to disable.
3. 3 Choose a stacking direction: Set the direction to above, below, or both. Both is the classic Zalgo look; one-sided stacks are useful when you want the source to stay readable.
4. 4 Set an optional seed: Enter a seed number so the same combination always produces the same styled text. Re-use a seed for an output you like.
5. 5 Copy the glitched text: Select the styled result and copy it with your usual shortcut, then paste it into a chat, a username, a comment box, or a social media post.

For a Halloween caption, type 'trick or treat', set intensity to 8, leave direction on both, and use seed 13. The result is a dramatic Zalgo string that you can paste into the social media field of your choice.

If you want to see how the same base characters look in a different machine encoding, the Text to Binary Converter turns the same source string into 8-bit binary and counts the output bytes alongside it.

A Zalgo generator is most useful when the styled text is fast to make, easy to copy, and predictable enough to repeat.

• • Faster than manual stacking: Replaces inserting diacritics one by one from a character map with a form submission.
• • Deterministic seed control: Lock in a favorite result by re-using the same seed, so a creative project stays consistent.
• • Three directions, one tool: Above, below, and both directions are available without leaving the page.
• • Counts you can trust: Shows the base character count and total combining mark count, so you can tell at a glance how heavy the effect is.
• • Plain preview included: Strips combining marks in a separate preview row, handy for verifying the original text was read correctly.

The plain preview is also useful as a fallback if a receiving app strips combining marks. You always have a clean version of the same text ready to copy if the styled version does not survive the trip through an older form field.

For repeated use, choose a small set of seeds and intensities that match the platforms you post on. Different apps cap stacking at different heights, so a single favorite seed will not look identical everywhere, but it will be close enough.

If you also want to know how the seed integer itself maps into other number bases, the Base Converter shows the same seed as binary, hex, and octal output so you can pick a value that reads well across formats.

The visible result depends on the source characters, the direction, the receiving app, and the platform limits on stacking.

• • The calculator does not insert Zalgo-style ligatures or replace base characters; it only stacks combining marks, so the underlying text stays readable in the plain preview.
• • According to the Unicode Consortium's Unicode Standard Annex #15 on Normalization Forms, every combining mark carries a canonical combining class that controls stacking order, so two seeds always layer predictably even when they pick the same marks in a different order.
• • Random sampling is pseudo-random and seeded, so two different seeds can produce visually similar but byte-different styled strings for the same source text.

Receiving app support is the most common reason a styled result looks weaker after pasting than it did in the result panel. If the field is set to only accept a small Unicode subset, the marks are removed before storage and the visible text falls back to the plain preview.

Stacking limits also matter if you plan to share the result on a phone. Mobile keyboards, accessibility tools, and screen readers handle moderate stacking fine but may struggle when a single line is hundreds of code points tall.

According to the Unicode Consortium's Unicode Standard Annex #15 on Normalization Forms, every combining mark carries a canonical combining class that controls the order in which marks stack on a single base character.