LaTeX to Unicode Converter
Paste LaTeX and get plain Unicode text — Greek letters, operators, superscripts and subscripts, blackboard-bold letters, accents, fractions and roots — ready for an email, a slide, a chat message or any place without an equation editor. Switch direction to turn Unicode symbols back into LaTeX commands, wrapped in dollar signs, with the packages you need listed. It runs in your browser — nothing is uploaded. Free, no sign-up.
Converted in your browser as you type. Nothing is uploaded.
Options
Enter some text, or load the example, to convert it.
How to Use the LaTeX to Unicode Converter
- 1. Pick a direction. LaTeX → Unicode turns commands into characters; Unicode → LaTeX does the reverse. Load the example shows a short passage either way.
- 2. Paste your text. A paragraph with
$…$formulas works, and so does a bare fragment such as\alpha_1 + \beta^2. The result updates as you type. - 3. Set the options. Going to Unicode, you can keep or drop the dollar signs, read LaTeX quote and dash ligatures as curly quotes and dashes, and strip
%comments. Going to LaTeX, you choose whether maths symbols are wrapped in dollar signs and whether special characters are escaped. - 4. Read the notes, then copy. The tool lists every command it could not convert and every superscript or subscript it left alone. For LaTeX output it also lists the
\usepackagelines to add to your preamble. Use the result as input and swap direction checks a round trip in one click.
What Converts to What
The symbol commands come from the same data as the LaTeX Symbol Finder, so a command that finder lists converts here and the reverse. The rest are rules, summarised below.
| LaTeX | Unicode | How it works |
|---|---|---|
\alpha, \Omega, \leq, \to | α Ω ≤ → | Every argument-free command in the symbol list: Greek letters, operators, relations, arrows, set and logic symbols. |
x^2, x_i, n^x | x² xᵢ nˣ | Digits, plus, minus, equals, brackets and most Latin letters have raised forms; fewer have lowered ones. A braced group is converted whole or not at all. |
\mathbb{R}, \mathcal{L}, \mathfrak{g}, \mathbf{x} | ℝ ℒ 𝔤 𝐱 | Unicode's mathematical alphabets. Letters convert in every style; digits only in bold, double-struck, sans-serif and monospace. |
\'e, \"o, \c{c}, \ss | é ö ç ß | The text accents and letter commands. A base letter and its accent are joined into one character wherever Unicode has one. |
\hat{x}, \bar{x}, \vec{v} | x̂ x̄ v⃗ | Math accents become combining marks placed after the letter. Fonts differ in how well they position them. |
\frac{1}{2}, \frac{a+b}{c} | ½ (a+b)/c | Fractions with a Unicode character use it; others become a slash, with brackets where the parts are not a single token. |
\sqrt{x+1}, \sqrt[3]{8} | √(x+1) ∛8 | Square, cube and fourth roots have their own signs; other indexes are raised in front of the root sign. |
~, ---, -- | no-break space, —, – | Read as TeX reads them in text. Inside a formula a hyphen becomes a true minus sign, −. |
A Worked Example
The example for LaTeX → Unicode is three lines. The first, The photon energy is $E = h\nu$ and its wavelength is $\lambda = \frac{hc}{E}$., becomes “The photon energy is E = hν and its wavelength is λ = hc/E.” The second line turns $e^x \geq 1 + x + \frac{x^2}{2}$ into eˣ ≥ 1 + x + x²/2, and the third turns \hat{\theta}_n \to \theta_0 into θ̂ₙ → θ₀, with \hat{\theta} as θ plus a combining circumflex and the subscripts as ₙ and ₀. The whole passage converts 20 items and leaves nothing behind.
Going the other way, “x² ≤ y² when x ≥ 0” becomes $x^2 \leq y^2$ when $x \geq 0$: the two formulas are separate because the words between them are text. “H₂O” becomes H$_2$O, because a word with a subscript in it is not a formula, while “m²” becomes $m^2$, because a single variable with an exponent is. Several raised characters, such as the minus and digit of ⁻³, share one braced group.
How the Reverse Direction Decides What Is Maths
Going to LaTeX, the tool has to guess which characters belong in maths mode, because Unicode does not say. Greek letters, operators, relations, arrows, superscripts, subscripts and mathematical-alphabet letters always do. The plain digits, variables and operators beside them join the same formula when they sit between two maths symbols, or form an operand such as the 5 and the 3 in 5 × 3. A gap with a word of two or more letters ends the formula. Typographic symbols such as ©, °, … and curly quotes stay in text mode, and accented letters become text accents (caf\'{e}) unless they are part of a formula, where they become math accents.
The result is a good first draft rather than a verdict. Read it over before compiling, and untick the dollar-sign option if you would rather place the delimiters yourself.
Packages the output may need
\mathbb and \mathfrak come from amssymb, many relations and arrows from amssymb or amsmath, and an ogonek such as ą needs the T1 font encoding. The tool lists these as \usepackage lines. Symbols from the textcomp set, such as the degree sign and the euro, are built into LaTeX since 2020; on an older installation, add \usepackage{textcomp}.
What the Converter Does Not Do
- It is not a TeX engine. Macros you define with
\newcommandare not expanded, so commands it does not know are kept exactly as typed and listed. - It cannot flatten two-dimensional maths. Matrices, aligned equations, stacked fractions and big operators with limits above and below have no honest one-line form. The
equationandalignwrappers are removed and their contents converted where they can be; a matrix's begin and end lines are kept and listed. - Plain text has no bold or italic.
\textbfand\emphkeep their words and lose their style;\mathbfuses the bold mathematical alphabet instead. - Some scripts are missing. Unicode has no raised q and no lowered b, c, d, f, g, q, w, y or z, or a lowered comma, so an index written with a comma stays as LaTeX. A group containing one is kept as typed, or written in brackets if you choose.
- Unicode maths is for reading, not for typesetting or accessibility. Screen readers announce a mathematical bold x as a long name, and fonts vary. For a paper, thesis or accessible document, keep the LaTeX or MathML source.
Frequently Asked Questions
- How do I convert LaTeX to plain text?
- Paste it into the converter with the direction set to LaTeX → Unicode. Commands become characters (
\alphabecomes α), exponents and indexes become raised and lowered characters, and formatting-only commands such as\leftand\displaystyleare dropped. Copy the result into any editor that accepts Unicode. - Can Unicode show every superscript and subscript?
- No. Unicode covers the digits 0–9, plus, minus, equals and brackets, most lowercase Latin letters for superscripts (not q), a smaller set of capitals, and a few Greek letters. Subscripts are more limited: only a, e, h, i, j, k, l, m, n, o, p, r, s, t, u, v, x and a few Greek letters. When a group needs a character outside those sets the converter leaves it as LaTeX instead of half-converting it.
- Why does my hat or bar accent look misplaced?
- Accents on letters like x have no single Unicode character, so they are written as a combining mark after the letter, and the font decides where to draw it. A maths font such as STIX Two Math or Cambria Math positions them best; in many text fonts the mark sits slightly off. Letters with a precomposed form, such as ā or ñ, are joined into one character and do not have this problem.
- Does it handle matrices and aligned equations?
- Not as layout. A matrix or an aligned block cannot be written on one line without losing its meaning, so its begin and end lines are kept as typed and listed. Everything inside an equation, align or similar maths environment that has a Unicode form is converted, and the environment's begin and end lines are removed.
- What happens to commands it does not know?
- They stay exactly as you typed them, together with the arguments right after them, and appear in the list of unconverted commands. Nothing is guessed. If a command is a symbol that should convert, check its spelling in the LaTeX Symbol Finder.
- Will the Unicode result compile if I paste it into a LaTeX document?
- With XeLaTeX or LuaLaTeX and a font that has the characters, text such as accented letters and Greek letters will print. In pdfLaTeX, characters outside the Latin script usually stop the build. That is the reason for the Unicode → LaTeX direction, which writes the commands back.
- Is my text uploaded anywhere?
- No. The conversion is plain JavaScript that runs in your browser; nothing you paste is sent to CitationEasy or anywhere else.