MediaWiki uses a subset of AMS-LaTeX markup, a superset of LaTeX markup which is in turn a superset of TeX markup, for mathematical formulae. It generates PNG images by default. Alternatively, the MathJax renderer can be selected in the user preferences: this uses a combination of HTML and CpageSS to display the equation.

Although, in all cases mentioned, Template:TeX is generated by compilation, and not by an interpreter program, there is one essential difference between, e.g., Knuth's TeX or Lamport's LaTeX and the present implementation: whereas in the first two cases the compiler typically generates an all-in-one printable output, which has the quality of a whole book with all chapters, sections and subsections, and where no line is "special", in the present case one has, typically, a mixture of Template:TeX images (more precisely: PNG images) for the equations, embedded into usual text, and with short Template:TeX elements usually replaced by HTML parts. As a consequence, in many cases TeX-elements, e.g. vector symbols, "stick out" below (or above) the text line. This "sticking out" is not  the case in the above-mentioned original products, and the HTML-substitutes for small Template:TeX additions to the text are often insufficient in quality for many readers. In spite of these shortcomings, the present product characterized by "many embedded PNG-images" should be preferred for small texts, where the equations do not dominate.

More precisely, MediaWiki filters the markup through Texvc, which in turn passes the commands to Template:TeX or MathJax for the actual rendering. Thus, only a limited part of the full Template:TeX language is supported; see below for details.

To have math rendered in a particular MediaWiki installation, one has to set $wgUseTeX = true; in LocalSettings.php.

Basics edit

Math markup goes inside ‎<math>...‎</math>.

The Template:TeX code has to be put literally: MediaWiki templates, predefined templates, and parameters cannot be used within math tags: pairs of double braces are ignored and "#" gives an error message. However, math tags work in the then and else part of #if, etc. See m:Template:Demo of attempt to use parameters within TeX for more information.

LaTeX commands edit

LaTeX commands are case-sensitive, and take one of the following two formats:

  • They start with a backslash \ and then have a name consisting of letters only. Command names are terminated by a space, a number or any other "non-letter".
  • They consist of a backslash \ and exactly one non-letter.

Some commands need an argument, which has to be given between curly braces { } after the command name. Some commands support optional parameters, which are added after the command name in square brackets []. The general syntax is:


Special characters edit

The following symbols are reserved characters that either have a special meaning under LaTeX or are unavailable in all the fonts. If you enter them directly in your text, they will normally not render, but rather do things you did not intend.

# $ % ^ & _ { } ~ \

These characters can be entered by adding a prefix backslash:

\# \$ \% \textasciicircum{} \& \_ \{ \} \~{} \textbackslash{}

The other symbols and many more can be rendered with special commands in mathematical formulae or as accents.

The backslash character \ can not be entered by adding another backslash in front of it (\\); this sequence is used for line breaking. For introducing a backslash in math mode, you can use \backslash instead.

The command \~ produces a tilde which is placed over the next letter. For example \~n gives ñ. To produce just the character ~, use \~{} which places a ~ over an empty box. Alternatively \sim produces a large centred ~ which may be more appropriate in some situations, but may not render properly in simple expressions which are converted to HTML.

Similarly, the command \^ produces a hat over the next character, for example \^{o} produces ô. If you need in text to display the ^ symbol you have to use \textasciicircum.

Spaces edit

"Whitespace" characters, such as blank or tab, are treated uniformly as "space" by LaTeX. Several consecutive whitespace characters are treated as one "space". See below for commands that produces spaces of different size.

LaTeX environments edit

Environments in LaTeX have a role that is quite similar to commands, but they usually have effect on a wider part of formula. Their syntax is:

   text to be influenced

Environments supported by Wikipedia include matrix, align, etc. See below.

Rendering edit

The colors, as well as font sizes and types, are independent of browser settings or CSS. Font sizes and types will often deviate from what HTML renders. Vertical alignment with the surrounding text can also be a problem (see bug 32694); a work-around is described in the "Alignment with normal text flow" section below. The css selector of the images is img.tex.

The alt text of the PNG images, which is displayed to visually impaired and other readers who cannot see the images, and is also used when the text is selected and copied, defaults to the wikitext that produced the image, excluding the <math> and </math>. You can override this by explicitly specifying an alt attribute for the math element. For example, <math alt="Square root of pi">\sqrt{\pi}</math> generates an image whose alt text is "Square root of pi". This should not be confused with the title attribute that produces popup text when the hovering over the PNG image, for example <math title="pi">\pi</math> generates an image whose popup text is "pi".

Apart from function and operator names, as is customary in mathematics, variables and letters are in italics; digits are not. For other text, (like variable labels) to avoid being rendered in italics like variables, use \text, \mbox, or \mathrm. You can also define new function names using \operatorname{...}. For example, \text{abc} gives .

Latex does not have full support for Unicode characters and not all characters render. Most Latin characters with accents render correctly. However some do not, in particular those that include multiple diacritics (e.g. with Latin letters used in Vietnamese) or that cannot be precomposed into a single character (such as the uppercase Latin letter W with ring), or that use other diacritics (like the ogonek or the double grave accent, used in Central European languages like Polish, or the horn attached above some vowels in Vietnamese), or other modified letter forms (used in IPA notations, or African languages, or in medieval texts), some digram ligatures (like IJ in Dutch), or Latin letters borrowed from Greek, or small capitals, as well as superscripts and subscript letters. For example \text{ð} or \mbox{ð}, and \text{þ} or \mbox{þ} (used in Icelandic) will give errors.

Template:TeX vs HTML edit

Before using Template:TeX markup for producing special characters, it should be noted that, as this comparison table shows, sometimes similar results can be achieved in HTML, with Template:Math or Template:Bigmath (also see: Help:Special characters).

Template:TeX syntax Template:TeX rendering HTML syntax HTML rendering
\alpha\,\! {{math|<VAR>&alpha;</VAR>}} Template:Math
f(x) = x^2\,\! {{math|''f''(<var>x</var>) {{=}} <var>x</var><sup>2</sup>}} Template:Math
\sqrt{2} {{math|{{radical|2}}}} Template:Math
\sqrt{1-e^2} {{math|{{radical|1 &minus; ''e''<sup>2</sup>}}}} Template:Math
\sqrt{1-z^3} {{bigmath|{{radical|1 &minus; ''z''<sup>3</sup>}}}} Template:Bigmath

The codes on the left produce the symbols on the right, but the latter can also be put directly in the wikitext, except for ‘=’.

Syntax Rendering
&alpha; &beta; &gamma; &delta; &epsilon; &zeta;
&eta; &theta; &iota; &kappa; &lambda; &mu; &nu;
&xi; &omicron; &pi; &rho; &sigma; &sigmaf;
&tau; &upsilon; &phi; &chi; &psi; &omega;

α β γ δ ε ζ
η θ ι κ λ μ ν
ξ ο π ρ σ ς
τ υ φ χ ψ ω

&Gamma; &Delta; &Theta; &Lambda; &Xi; &Pi;
&Sigma; &Phi; &Psi; &Omega;


&int; &sum; &prod; &radic; &minus; &plusmn; &infin;
&asymp; &prop; = &equiv; &ne; &le; &ge;
&times; &middot; &sdot; &divide; &part; &prime; &Prime;
&nabla; &permil; &deg; &there4; &empty;

∫ ∑ ∏ √ − ± ∞
≈ ∝ = ≡ ≠ ≤ ≥
× · ⋅ ÷ ∂ ′ ″
∇ ‰ ° ∴ ∅

&isin; &notin; &cap; &cup; &sub; &sup; &sube; &supe;
&not; &and; &or; &exist; &forall;
&rArr; &hArr; &rarr; &harr; &uarr; &darr;
&alefsym; - &ndash; &mdash;

∈ ∉ ∩ ∪ ⊂ ⊃ ⊆ ⊇
¬ ∧ ∨ ∃ ∀
⇒ ⇔ → ↔ ↑ ↓
ℵ - – —

The project has settled on using both HTML and Template:TeX because each has advantages in some situations.

Pros of HTML edit

  1. Formulas in HTML behave more like regular text. In-line HTML formulae always align properly with the rest of the HTML text and, to some degree, can be copied-and-pasted (this is not a problem if Template:TeX is rendered using MathJax, and the alignment should not be a problem for PNG rendering once bug 32694 is fixed).
  2. Pages using HTML code for formulae will load faster and they will create less clutter on your hard disk.
  3. Formulae typeset with HTML code will be accessible to client-side script links (a.k.a. scriptlets).
  4. The display of a formula entered using mathematical templates can be conveniently altered by modifying the templates involved; this modification will affect all relevant formulae without any manual intervention.
  5. The HTML code, if entered diligently, will contain all semantic information to transform the equation back to Template:TeX or any other code as needed. It can even contain differences Template:TeX does not normally catch, e.g. {{math|''i''}} for the imaginary unit and {{math|<var>i</var>}} for an arbitrary index variable.
  6. Unlike generated bitmaps, HTML is not sensitive to dots per inch variances between viewing platforms.

Pros of Template:TeX edit

  1. Template:TeX is semantically more precise than HTML.
    1. In Template:TeX, "x" means "mathematical variable ", whereas in HTML "x" is generic and somewhat ambiguous.
    2. On the other hand, if you encode the same formula as "{{math|<var>x</var>}}", you get the same visual result Template:Math and no information is lost. This requires diligence and more typing that could make the formula harder to understand as you type it. However, since there are far more readers than editors, this effort is worth considering if no other rendering options are available (such as MathJax, which is available to logged in users as an preferences opt-in).
  2. One consequence of point 1 is that Template:TeX code can be transformed into HTML, but not vice-versa.Template:Ref This means that on the server side we can always transform a formula, based on its complexity and location within the text, user preferences, type of browser, etc. Therefore, where possible, all the benefits of HTML can be retained, together with the benefits of Template:TeX. It is true that the current situation is not ideal, but that is not a good reason to drop information/contents. It is more a reason to help improve the situation.
  3. Another consequence of point 1 is that Template:TeX can be converted to MathML (e.g. by MathJax) for browsers which support it, thus keeping its semantics and allowing the rendering to be better suited for the reader’s graphic device.
  4. Template:TeX is the preferred text formatting language of most professional mathematicians, scientists, and engineers. It is easier to persuade them to contribute if they can write in Template:TeX.
  5. Template:TeX has been specifically designed for typesetting formulae, so input is easier and more natural if you are accustomed to it, and output is more aesthetically pleasing if you focus on a single formula rather than on the whole containing page.
  6. Once a formula is done correctly in Template:TeX, it will render reliably, whereas the success of HTML formulae is somewhat dependent on browsers or versions of browsers. Another aspect of this dependency is fonts: the serif font used for rendering formulae is browser-dependent and it may be missing some important glyphs. While the browser generally capable to substitute a matching glyph from a different font family, it need not be the case for combined glyphs (compare ‘  ’ and ‘  ’).
  7. When writing in Template:TeX, editors need not worry about whether this or that version of this or that browser supports this or that HTML entity. The burden of these decisions is put on the software. This does not hold for HTML formulae, which can easily end up being rendered wrongly or differently from the editor’s intentions on a different browser.Template:Ref
  8. Template:TeX formulae, by default, render larger and are usually more readable than HTML formulae and are not dependent on client-side browser resources, such as fonts, and so the results are more reliably WYSIWYG.
  9. While Template:TeX does not assist you in finding HTML codes or Unicode values (which you can obtain by viewing the HTML source in your browser), copying and pasting from a Template:TeX PNG image in Wikipedia into simple text will return the LaTeX source.
unless your wikitext follows the style of point 1.2
The entity support problem is not limited to mathematical formulae though; it can be easily solved by using the corresponding characters instead of entities, as the character repertoire links do, except for cases where the corresponding glyphs are visually indiscernible (e.g. &ndash; for ‘–’ and &minus; for ‘−’).

In some cases it may be the best choice to use neither Template:TeX nor the HTML substitutes, but instead the simple ASCII symbols of a standard keyboard (see hereafter, for an example).

Using MathJax edit

The MathJax renderer, selectable through My Preferences - Appearance - Math, uses a very different system to the standard texvc renderer. Rather than rendering a static image on the server side a combination of JavaScript, HTML and CSS is used to locally construct the formula. This allows high-quality typesetting, and other problems such as font sizes not matching your browser settings or wrong baselines are fixed as well. MathJax may also reduce the download time of images but entails a small delay as the JavaScript interpreter runs to render the formulas. (After the small delay, the vertical dimension of the displayed page tends to update asynchronously—math content usually becomes taller—which can cause a disorienting jump in the document view position in some user agents; speedy mouse-wheelers might wish to check into this as part of their usability evaluation.)

Alternatively to the preferences option, MathJax can also be selected using the mathJax user script. It is the basis for the global option, and hence more experimental but also more up to date. Bug reports are taken care of at the script's talk page.

The quality of font rendering is dependent on your browser and operating system. Speed and appearance can be improved by installing the STIX fonts; for installation details see the MathJax font help (Archived 2012-07-28 at the Wayback Machine) page. Further information is also available at this page.

Formatting using Template:TeX edit

Functions, symbols, special characters edit

Accents/diacritics edit

\dot{a}, \ddot{a}, \acute{a}, \grave{a}  
\check{a}, \breve{a}, \tilde{a}, \bar{a}  
\hat{a}, \widehat{a}, \vec{a}  

Standard numerical functions edit

\exp_a b = a^b, \exp b = e^b, 10^m  
\ln c, \lg d = \log e, \log_{10} f  
\sin a, \cos b, \tan c, \cot d, \sec e, \csc f  
\arcsin h, \arccos i, \arctan j  
\sinh k, \cosh l, \tanh m, \coth n  
\operatorname{sh}\,k, \operatorname{ch}\,l, \operatorname{th}\,m, \operatorname{coth}\,n  
\operatorname{argsh}\,o, \operatorname{argch}\,p, \operatorname{argth}\,q  
\sgn r, \left\vert s \right\vert  
\min(x,y), \max(x,y)  

Bounds edit

\min x, \max y, \inf s, \sup t  
\lim u, \liminf v, \limsup w  
\dim p, \deg q, \det m, \ker\phi  

Projections edit

\Pr j, \hom l, \lVert z \rVert, \arg z  

Differentials and derivatives edit

dt, \operatorname{d}\!t, \partial t, \nabla\psi  
dy/dx, \operatorname{d}\!y/\operatorname{d}\!x, {dy \over dx}, {\operatorname{d}\!y\over\operatorname{d}\!x}, {\partial^2\over\partial x_1\partial x_2}y  
\prime, \backprime, f^\prime, f', f'', f^{(3)}, \dot y, \ddot y  

Letter-like symbols or constants edit

\infty, \aleph, \complement, \backepsilon, \eth, \Finv, \hbar  
\Im, \imath, \jmath, \Bbbk, \ell, \mho, \wp, \Re, \circledS  

Modular arithmetic edit

s_k \equiv 0 \pmod{m}  
\gcd(m, n), \operatorname{lcm}(m, n)  
\mid, \nmid, \shortmid, \nshortmid  

Radicals edit

\surd, \sqrt{2}, \sqrt[n]{}, \sqrt[3]{x^3+y^3 \over 2}  

Operators edit

+, -, \pm, \mp, \dotplus  
\times, \div, \divideontimes, /, \backslash  
\cdot, * \ast, \star, \circ, \bullet  
\boxplus, \boxminus, \boxtimes, \boxdot  
\oplus, \ominus, \otimes, \oslash, \odot  
\circleddash, \circledcirc, \circledast  
\bigoplus, \bigotimes, \bigodot  

Sets edit

\{ \}, \O \empty \emptyset, \varnothing  
\in, \notin \not\in, \ni, \not\ni  
\cap, \Cap, \sqcap, \bigcap  
\cup, \Cup, \sqcup, \bigcup, \bigsqcup, \uplus, \biguplus  
\setminus, \smallsetminus, \times  
\subset, \Subset, \sqsubset  
\supset, \Supset, \sqsupset Failed to parse (SVG (MathML can be enabled via browser plugin): Invalid response ("Math extension cannot connect to Restbase.") from server "http://localhost:6011/www.mediawiki.org/v1/":): {\displaystyle \supset, \Supset, \sqsupset \!}
\subseteq, \nsubseteq, \subsetneq, \varsubsetneq, \sqsubseteq  
\supseteq, \nsupseteq, \supsetneq, \varsupsetneq, \sqsupseteq  
\subseteqq, \nsubseteqq, \subsetneqq, \varsubsetneqq  
\supseteqq, \nsupseteqq, \supsetneqq, \varsupsetneqq  

Relations edit

=, \ne, \neq, \equiv, \not\equiv  
\doteq, \doteqdot, \overset{\underset{\mathrm{def}}{}}{=}, :=  
\sim, \nsim, \backsim, \thicksim, \simeq, \backsimeq, \eqsim, \cong, \ncong  
\approx, \thickapprox, \approxeq, \asymp, \propto, \varpropto  
<, \nless, \ll, \not\ll, \lll, \not\lll, \lessdot  
>, \ngtr, \gg, \not\gg, \ggg, \not\ggg, \gtrdot  
\le, \leq, \lneq, \leqq, \nleq, \nleqq, \lneqq, \lvertneqq  
\ge, \geq, \gneq, \geqq, \ngeq, \ngeqq, \gneqq, \gvertneqq  
\lessgtr, \lesseqgtr, \lesseqqgtr, \gtrless, \gtreqless, \gtreqqless  
\leqslant, \nleqslant, \eqslantless  
\geqslant, \ngeqslant, \eqslantgtr  
\lesssim, \lnsim, \lessapprox, \lnapprox  
\gtrsim, \gnsim, \gtrapprox, \gnapprox  
\prec, \nprec, \preceq, \npreceq, \precneqq  
\succ, \nsucc, \succeq, \nsucceq, \succneqq  
\preccurlyeq, \curlyeqprec  
\succcurlyeq, \curlyeqsucc  
\precsim, \precnsim, \precapprox, \precnapprox  
\succsim, \succnsim, \succapprox, \succnapprox  

Geometric edit

\parallel, \nparallel, \shortparallel, \nshortparallel  
\perp, \angle, \sphericalangle, \measuredangle, 45^\circ  
\Box, \blacksquare, \diamond, \Diamond \lozenge, \blacklozenge, \bigstar  
\bigcirc, \triangle \bigtriangleup, \bigtriangledown  
\vartriangle, \triangledown  
\blacktriangle, \blacktriangledown, \blacktriangleleft, \blacktriangleright  

Logic edit

\forall, \exists, \nexists  
\therefore, \because, \And  
\or \lor \vee, \curlyvee, \bigvee  
\and \land \wedge, \curlywedge, \bigwedge  
\bar{q}, \bar{abc}, \overline{q}, \overline{abc},

\lnot \neg, \not\operatorname{R}, \bot, \top



\vdash \dashv, \vDash, \Vdash, \models  
\Vvdash \nvdash \nVdash \nvDash \nVDash  
\ulcorner \urcorner \llcorner \lrcorner  

Arrows edit

\Rrightarrow, \Lleftarrow  
\Rightarrow, \nRightarrow, \Longrightarrow \implies  
\Leftarrow, \nLeftarrow, \Longleftarrow  
\Leftrightarrow, \nLeftrightarrow, \Longleftrightarrow \iff  
\Uparrow, \Downarrow, \Updownarrow  
\rightarrow \to, \nrightarrow, \longrightarrow  
\leftarrow \gets, \nleftarrow, \longleftarrow  
\leftrightarrow, \nleftrightarrow, \longleftrightarrow  
\uparrow, \downarrow, \updownarrow  
\nearrow, \swarrow, \nwarrow, \searrow  
\mapsto, \longmapsto  
\rightharpoonup \rightharpoondown \leftharpoonup \leftharpoondown \upharpoonleft \upharpoonright \downharpoonleft \downharpoonright \rightleftharpoons \leftrightharpoons  
\curvearrowleft \circlearrowleft \Lsh \upuparrows \rightrightarrows \rightleftarrows \rightarrowtail \looparrowright  
\curvearrowright \circlearrowright \Rsh \downdownarrows \leftleftarrows \leftrightarrows \leftarrowtail \looparrowleft  
\hookrightarrow \hookleftarrow \multimap \leftrightsquigarrow \rightsquigarrow \twoheadrightarrow \twoheadleftarrow  

Special edit

\amalg \P \S \% \dagger \ddagger \ldots \cdots  
\smile \frown \wr \triangleleft \triangleright  
\diamondsuit, \heartsuit, \clubsuit, \spadesuit, \Game, \flat, \natural, \sharp  

Unsorted (new stuff) edit

\diagup \diagdown \centerdot \ltimes \rtimes \leftthreetimes \rightthreetimes  
\eqcirc \circeq \triangleq \bumpeq \Bumpeq \doteqdot \risingdotseq \fallingdotseq  
\intercal \barwedge \veebar \doublebarwedge \between \pitchfork  
\vartriangleleft \ntriangleleft \vartriangleright \ntriangleright  
\trianglelefteq \ntrianglelefteq \trianglerighteq \ntrianglerighteq  

For a little more semantics on these symbols, see the brief TeX Cookbook.

Larger expressions edit

Subscripts, superscripts, integrals edit

Feature Syntax How it looks rendered
Superscript a^2  
Subscript a_2  
Grouping 10^{30} a^{2+2}  
a_{i,j} b_{f'}  
Combining sub & super without and with horizontal separation x_2^3  
Super super 10^{10^{8}}  
Preceding and/or additional sub & super \sideset{_1^2}{_3^4}\prod_a^b  
Stacking \overset{\alpha}{\omega}  
Derivatives x', y'', f', f''  
x^\prime, y^{\prime\prime}  
Derivative dots \dot{x}, \ddot{x}  
Underlines, overlines, vectors \hat a \ \bar b \ \vec c  
\overrightarrow{a b} \ \overleftarrow{c d} \ \widehat{d e f}  
\overline{g h i} \ \underline{j k l}  
Arc (workaround) \overset{\frown} {AB}  
Arrows A \xleftarrow{n+\mu-1} B \xrightarrow[T]{n\pm i-1} C  
Overbraces \overbrace{ 1+2+\cdots+100 }^{5050}  
Underbraces \underbrace{ a+b+\cdots+z }_{26}  
Sum \sum_{k=1}^N k^2  
Sum (force \textstyle) \textstyle \sum_{k=1}^N k^2  
Sum in a fraction (default \textstyle) \frac{\sum_{k=1}^N k^2}{a}  
Sum in a fraction (force \displaystyle) \frac{\displaystyle \sum_{k=1}^N k^2}{a}  
Sum in a fraction (alternative limits style) \frac{\sum\limits^{^N}_{k=1} k^2}{a}  
Product \prod_{i=1}^N x_i  
Product (force \textstyle) \textstyle \prod_{i=1}^N x_i  
Coproduct \coprod_{i=1}^N x_i  
Coproduct (force \textstyle) \textstyle \coprod_{i=1}^N x_i  
Limit \lim_{n \to \infty}x_n  
Limit (force \textstyle) \textstyle \lim_{n \to \infty}x_n  
Integral \int\limits_{1}^{3}\frac{e^3/x}{x^2}\, dx  
Integral (alternative limits style) \int_{1}^{3}\frac{e^3/x}{x^2}\, dx  
Integral (force \textstyle) \textstyle \int\limits_{-N}^{N} e^x\, dx  
Integral (force \textstyle, alternative limits style) \textstyle \int_{-N}^{N} e^x\, dx  
Double integral \iint\limits_D \, dx\,dy  
Triple integral \iiint\limits_E \, dx\,dy\,dz  
Quadruple integral \iiiint\limits_F \, dx\,dy\,dz\,dt  
Line or path integral \int_{(x,y)\in C} x^3\, dx + 4y^2\, dy  
Closed line or path integral \oint_{(x,y)\in C} x^3\, dx + 4y^2\, dy  
Intersections \bigcap_{i=_1}^n E_i  
Unions \bigcup_{i=_1}^n E_i  

Fractions, matrices, multilines edit

Feature Syntax How it looks rendered
Fractions \frac{2}{4}=0.5 or {2 \over 4}=0.5  
Small fractions \tfrac{2}{4} = 0.5  
Large (normal) fractions \dfrac{2}{4} = 0.5 \qquad \dfrac{2}{c + \dfrac{2}{d + \dfrac{2}{4}}} = a  
Large (nested) fractions \cfrac{2}{c + \cfrac{2}{d + \cfrac{2}{4}}} = a  
Cancellations in fractions
Note does not work with MathJax
\cfrac{x}{1 + \cfrac{\cancel{y}}{\cancel{y}}} = \cfrac{x}{2}  
Binomial coefficients \binom{n}{k}  
Small binomial coefficients \tbinom{n}{k}  
Large (normal) binomial coefficients \dbinom{n}{k}  
x & y \\
z & v
x & y \\
z & v
x & y \\
z & v
0 & \cdots & 0 \\
\vdots & \ddots & \vdots \\
0 & \cdots & 0
x & y \\
z & v
x & y \\
z & v
\bigl( \begin{smallmatrix}
a&b\\ c&d
\end{smallmatrix} \bigr)
Case distinctions
f(n) =
n/2, & \text{if }n\text{ is even} \\
3n+1, & \text{if }n\text{ is odd}
Multiline equations
f(x) & = (a+b)^2 \\
& = a^2+2ab+b^2 \\
f(x) & = (a-b)^2 \\
& = a^2-2ab+b^2 \\
Multiline equations (must define number of columns used ({lcr}) (should not be used unless needed)
z & = & a \\
f(x,y,z) & = & x + y + z
Multiline equations (more)
z & = & a \\
f(x,y,z) & = & x + y + z
Breaking up a long expression so that it wraps when necessary, at the expense of destroying correct spacing
f(x) =
\sum_{n=0}^\infty a_n x^n =
Simultaneous equations
3x + 5y + z \\
7x - 2y + 4z \\
-6x + 3y + 2z
\begin{array}{|c|c||c|} a & b & S \\

Parenthesizing big expressions, brackets, bars edit

Feature Syntax How it looks rendered
Bad ( \frac{1}{2} )  
Good \left ( \frac{1}{2} \right )  

You can use various delimiters with \left and \right:

Feature Syntax How it looks rendered
Parentheses \left ( \frac{a}{b} \right )  
Brackets \left [ \frac{a}{b} \right ] \quad
\left \lbrack \frac{a}{b} \right \rbrack
Braces \left \{ \frac{a}{b} \right \} \quad
\left \lbrace \frac{a}{b} \right \rbrace
Angle brackets \left \langle \frac{a}{b} \right \rangle  
Bars and double bars \left | \frac{a}{b} \right \vert \quad
\left \Vert \frac{c}{d} \right \|
Floor and ceiling functions: \left \lfloor \frac{a}{b} \right \rfloor \quad
\left \lceil \frac{c}{d} \right \rceil
Slashes and backslashes \left / \frac{a}{b} \right \backslash  
Up, down, and up-down arrows \left \uparrow \frac{a}{b} \right \downarrow \quad
\left \Uparrow \frac{a}{b} \right \Downarrow \quad
\left \updownarrow \frac{a}{b} \right \Updownarrow
Delimiters can be mixed,
as long as \left and \right match
\left [ 0,1 \right )
\left \langle \psi \right |
Use \left. and \right. if you
do not want a delimiter to appear
\left . \frac{A}{B} \right \} \to X  
Size of the delimiters (add "l" or "r" to indicate the side for proper spacing) ( \bigl( \Bigl( \biggl( \Biggl( \dots \Biggr] \biggr] \Bigr] \bigr] ]  
\{ \bigl\{ \Bigl\{ \biggl\{ \Biggl\{ \dots
\Biggr\rangle \biggr\rangle \Bigr\rangle \bigr\rangle \rangle
\| \big\| \Big\| \bigg\| \Bigg\| \dots \Bigg| \bigg| \Big| \big| |  
\lfloor \bigl\lfloor \Bigl\lfloor \biggl\lfloor \Biggl\lfloor \dots
\Biggr\rceil \biggr\rceil \Bigr\rceil \bigr\rceil \ceil
\uparrow \big\uparrow \Big\uparrow \bigg\uparrow \Bigg\uparrow \dots
\Bigg\Downarrow \bigg\Downarrow \Big\Downarrow \big\Downarrow \Downarrow
\updownarrow \big\updownarrow \Big\updownarrow \bigg\updownarrow \Bigg\updownarrow \dots
\Bigg\Updownarrow \bigg\Updownarrow \Big\Updownarrow \big\Updownarrow \Updownarrow
/ \big/ \Big/ \bigg/ \Bigg/ \dots
\Bigg\backslash \bigg\backslash \Big\backslash \big\backslash \backslash

Equation numbering edit

The templates {{NumBlk }} and {{EquationRef }} can be used to number equations. The template {{EquationNote }} can be used to refer to a numbered equation from surrounding text. For example, the following syntax:

{{NumBlk|:|<math>x^2 + y^2 + z^2 = 1 \,</math>|{{EquationRef|1}}}}

produces the following result (note the equation number in the right margin):


Later on, the text can refer to this equation by its number using syntax like this:

As seen in equation ({{EquationNote|1}}), blah blah blah...

The result looks like this:

As seen in equation (Template:EquationNote), blah blah blah...

Note that the equation number produced by {{EquationNote }} is a link that the user can click to go immediately to the cited equation.

Alphabets and typefaces edit

Texvc cannot render arbitrary Unicode characters. Those it can handle can be entered by the expressions below. For others, such as Cyrillic, they can be entered as Unicode or HTML entities in running text, but cannot be used in displayed formulas.

Greek alphabet
\Alpha \Beta \Gamma \Delta \Epsilon \Zeta \Eta \Theta  
\Iota \Kappa \Lambda \Mu \Nu \Xi \Pi \Rho  
\Sigma \Tau \Upsilon \Phi \Chi \Psi \Omega  
\alpha \beta \gamma \delta \epsilon \zeta \eta \theta  
\iota \kappa \lambda \mu \nu \xi \pi \rho  
\sigma \tau \upsilon \phi \chi \psi \omega  
\varepsilon \digamma \varkappa \varpi  
\varrho \varsigma \vartheta \varphi  
Hebrew symbols
\aleph \beth \gimel \daleth  
Blackboard bold/scripts
Boldface (Greek)
Italics (default for Latin alphabet)
Greek italics (default for lowercase Greek)
Roman typeface
Sans serif
Sans serif Greek (capital only)
\mathsf{\Alpha \Beta \Gamma \Delta \Epsilon \Zeta \Eta \Theta}  
\mathsf{\Iota \Kappa \Lambda \Mu \Nu \Xi \Pi \Rho}  
\mathsf{\Sigma \Tau \Upsilon \Phi \Chi \Psi \Omega}  
Fraktur typeface
Small scriptstyle text

Mixed text faces edit

Feature Syntax How it looks rendered
Italicised characters (spaces are ignored) x y z  
Non-italicised characters \text{x y z}  
Mixed italics (bad) \text{if} n \text{is even}  
Mixed italics (good) \text{if }n\text{ is even}  
Mixed italics (alternative: ~ or "\ " forces a space) \text{if}~n\ \text{is even}  

Color edit

Equations can use color with the \color command. The default Texvc renderer and the MathJax renderers have different syntaxes to support both use {\color{Blue}{text}}. For example

  • {\color{Blue}{x^2}}+{\color{Orange}{2x}}-{\color{LimeGreen}{1}}
  • x_{1,2}=\frac{{\color{Blue}{-b}}\pm\sqrt{\color{Red}{b^2-4ac}}}{\color{Green}{2a}}

There are several alternate notations styles

  • {\color{Blue}x^2}+{\color{Orange}2x}-{\color{LimeGreen}1} works with both texvc and MathJax
  • \color{Blue}x^2\color{Black}+\color{Orange}2x\color{Black}-\color{LimeGreen}1 works with both texvc and MathJax
  • \color{Blue}{x^2}+\color{Orange}{2x}-\color{LimeGreen}{1} only works with MathJax

Some color names are predeclared according to the following table, you can use them directly for the rendering of formulas (or for declaring the intended color of the page background).

Colors supported

Note that color should not be used as the only way to identify something, because it will become meaningless on black-and-white media or for color-blind people. See Wikipedia:Manual of Style (accessibility)#Color.

Custom colours can be defined using

\definecolor{myorange}{RGB}{255,165,100}\color{myorange}e^{i \pi}\color{Black} + 1 = 0

Formatting issues edit

Spacing edit

Note that Template:TeX handles most spacing automatically, but you may sometimes want manual control.

Feature Syntax How it looks rendered
double quad space a \qquad b  
quad space a \quad b  
text space a\ b  
text space without PNG conversion a \mbox{ } b  
large space a\;b  
medium space a\<b [not supported]
small space a\,b  
tiny space (use for multiplication of factors) ab  
no space (use for multi-letter variables) \mathit{ab}  
small negative space a\!b  

Automatic spacing may be broken in very long expressions (because they produce an overfull hbox in Template:TeX):


This can be remedied by putting a pair of braces { } around the whole expression:


Alignment with normal text flow edit

Due to the default CSS

img.tex { vertical-align: middle; }

an inline expression like   should look good.

If you need to align it otherwise, use <math style="vertical-align:-100%;">...</math> and play with the vertical-align argument until you get it right; however, how it looks may depend on the browser and the browser settings.

Also note that if you rely on this workaround, if/when the rendering on the server gets fixed in future releases, as a result of this extra manual offset your formulae will suddenly be aligned incorrectly. So use it sparingly, if at all.

Commutative diagrams edit

To make a commutative diagram, there are three steps:

  1. write the diagram in TeX
  2. convert to SVG
  3. upload the file to Wikimedia Commons

Diagrams in Template:TeX edit

Xy-pic (online manual) is the most powerful and general-purpose diagram package in TeX. Diagrams created using it can be found at Commons: Category:Xy-pic diagrams.

Simpler packages include:

The following is a template for Xy-pic, together with a hack to increase the margins in dvips, so that the diagram is not truncated by over-eager cropping (suggested in TUGboat: TUGboat, Volume 17 1996, No. 3):

\usepackage[all, ps, dvips]{xy}  % Loading the XY-Pic package
                                 % Using postscript driver for smoother curves
\usepackage{color}  % For invisible frame
\thispagestyle{empty}  % No page numbers
\SelectTips{eu}{}  % Euler arrowheads (tips)
\setlength{\fboxsep}{0pt}  % Frame box margin
{\color{white}\framebox{{\color{black}$$  % Frame for margin

%%% Diagram goes here %%%

$$}}} % end math, end frame

Convert to SVG edit

Once you have produced your diagram in LaTeX (or TeX), you can convert it to an SVG file using the following sequence of commands:

pdflatex file.tex
pdfcrop --clip file.pdf tmp.pdf
pdf2svg tmp.pdf file.svg
rm tmp.pdf

The pdfcrop and pdf2svg utilities are needed for this procedure. You can alternatively use pdf2svg from PDFTron for the last step.

If you do not have pdfTeX (which is unlikely) you can use the following commands to replace the first step (TeX → PDF):

latex file.tex
dvipdfm file.dvi

In general, you will not be able to get anywhere with diagrams without Template:TeX and Ghostscript, and the inkscape program is a useful tool for creating or modifying your diagrams by hand. There is also a utility pstoedit which supports direct conversion from Postscript files to many vector graphics formats, but it requires a non-free plugin to convert to SVG, and regardless of the format, this editor has not been successful in using it to convert diagrams with diagonal arrows from TeX-created files.

These programs are:

Upload the file edit

As the diagram is your own work, upload it to Wikimedia Commons, so that all projects (notably, all languages) can use it without having to copy it to their language's Wiki. (If you've previously uploaded a file to somewhere other than Commons, to Commons.)

Check size
Before uploading, check that the default size of the image is neither too large nor too small by opening in an SVG application and viewing at default size (100% scaling), otherwise adjust the -y option to dvips.
Make sure the file has a meaningful name.
Login to Wikimedia Commons, then upload the file; for the Summary, give a brief description.

Now go to the image page and add a description, including the source code, using this template:

|description =
{{en|1= Description [[:en:Link to WP page|topic]]
|source = {{own}}, created as per:
 [[:en:Help:Displaying a formula#Commutative diagrams]];
 source code below.
|date = The Creation Date, like 1999-12-31
|author = [[User:YourUserName|Your Real Name]]
|permission = {{self|PD-self (or other license)
    |author = [[User:YourUserName|Your Real Name]]}}

==TeX source==
 '''% TeX source here'''
[[Category:Commutative diagrams]]
[[Category:Xy-pic diagrams]]
[[Category:Images with LaTeX source code]]
Source code
  • Include the source code in the image page, in the Source section of the {{Information}} template, so that the diagram can be edited in future.
  • Include the complete .tex file, not just the fragment, so future editors do not need to reconstruct a compilable file.
  • You may optionally make the source code section collapsible, using the {{cot}}/{{cob}} templates.
  • (Don't include it in the Summary section, which is just supposed to be a summary.)
The most common license for commutative diagrams is PD-self; some use PD-ineligible, especially for simple diagrams, or other licenses. Please do not use the GFDL, as it requires the entire text of the GFDL to be attached to any document that uses the diagram.
If possible, link to a Wikipedia page relevant to the diagram. (The 1= is necessary if you use nest templates within the description, and harmless otherwise.)
Include [[Category:Commutative diagrams]], so that it appears in commons:Category:Commutative diagrams. There are also subcategories, which you may choose to use.
Include image
Now include the image on the original page via [[File:Diagram.svg]]

Examples edit

A sample conforming diagram is commons:Image:PSU-PU.svg.

Unimplemented elements and workarounds edit

\oiint and \oiiint edit

Elements which are not yet implemented are \oiint, a two-fold integral \iint ( ) with a circular curve through the centre of the two integrals, and similarly \oiiint, a circular curve through three integrals. In contrast, \oint ( ) exists for the single dimension (integration over a curved line within a plane or any space with higher dimension).

These elements appear in many contexts: \oiint denotes a surface integral over the closed 2d boundary of a 3d region (which occurs in much of 3d vector calculus and physical applications – like Maxwell's equations), likewise \oiiint denotes integration over the closed 3d boundary (surface volume) of a 4d region, and they would be strong candidates for the next Template:TeX version. As such there are a lot of workarounds in the present version.

\oiint and \oiiint using currently implemented symbols
\oiint looks like:
  •  , which uses \iint along with \subset and \supset (overdrawn after backspacing):
\iint\limits_{S}\!\!\!\!\!\!\!\!\!\!\!\subset\!\supset \mathbf D \cdot \mathrm{d}\mathbf A
  •  , which uses \int twice (with some backward kerning) along with \bigcirc (also overdrawn after backpacing) to produce a more consistent circle:
\int\!\!\!\!\int_{\partial V}\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\;\;\;\bigcirc\,\,\mathbf D\cdot\mathrm{d}\mathbf A

\oiiint (should also be preferably more tightly kerned) looks more or less like:

  •   which uses three \int symbols (with more backward kerning) with \subset and \supset (overdrawn after backspacing):
\int\!\!\!\!\!\int\!\!\!\!\!\int_{\partial V}\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\;\;\;\subset\!\supset \mathbf D\;\cdot\mathrm{d}\mathbf A
  •  , which uses three \int symbols (with more backward kerning) along with \bigcirc (also overdrawn after backspacing):
\int\!\!\!\!\!\int\!\!\!\!\!\int_{\partial V}\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\!\;\;\;\bigcirc\,\,\mathbf D\;\cdot\mathrm{d}\mathbf A

However, since no standardisation exists as yet, any workaround like this (which uses many \! symbols for backspacing) should be avoided, if possible. See below for a possibility using PNG image enforcement.

Note that \iint (the double integral) and \iiint (the triple integral) are still not kerned as they should preferably be, and are currently rendered as if they were successive \int symbols ; this is not a major problem for reading the formulas, even if the integral symbols before the last one do not have bounds, so it's best to avoid backspacing "hacks" as they may be inconsistent with a possible future better implementation of integrals symbols (with more precisely computed kerning positions).

\oiint and \oiiint as PNG images edit

These symbols are available as PNG images which are also integrated into two templates, {{Oiint }} and {{Oiiint }}, which take care of the formatting around the symbols.

The templates have three parameters:

the text or formula immediately before the integral
the subscript below the integral
the text or formula immediately after the integral
Examples edit
  • Stokes' theorem: {{oiint | intsubscpt = <math>{\scriptstyle S}</math> | integrand=<math>( \nabla \times \bold{F} ) \cdot {\rm d}\bold{S} = \oint_{\partial S} \bold{F} \cdot {\rm d}\boldsymbol{\ell} </math> }}
  • Ampère's law + correction: {{oiint | preintegral=<math>\oint_C \bold{B} \cdot {\rm d} \boldsymbol{\ell} = \mu_0 </math> | intsubscpt = <math>{\scriptstyle S}</math> | integrand = <math>\left ( \bold{J} + \epsilon_0\frac{\partial \bold{E}}{\partial t} \right ) \cdot {\rm d}\bold{S}</math> }}
  • Continuity of 4-momentum flux (in general relativity):[1] {{oiiint | preintegral=<math>\bold{P} = </math> | intsubscpt = <math>{\scriptstyle \partial \Omega}</math> | integrand = <math>\bold{T} \cdot {\rm d}^3\boldsymbol{\Sigma} = 0</math> }}

Oriented \oiint and \oiiint as PNG images edit

Some variants of \oiint and \oiiint have arrows on them to indicate the sense of integration, such as a line integral around a closed curve in the clockwise sense, and higher dimensional analogues. These are not implemented in Template:TEX on Wikipedia either, although the template {{Intorient }} is available - see link for details.

\overarc edit

\overarc is not yet implemented to display the arc notation. However, there exists a workaround: use \overset{\frown}{AB}, which gives  

Examples of implemented Template:TeX formulas edit

Quadratic polynomial edit


<math>ax^2 + bx + c = 0</math>

Quadratic formula edit


<math>x={-b\pm\sqrt{b^2-4ac} \over 2a}</math>

Tall parentheses and fractions edit


<math>2 = \left(
\frac{\left(3-x\right) \times 2}{3-x}


<math>S_{\text{new}} = S_{\text{old}} - \frac{ \left( 5-T \right) ^2} {2}</math>

Integrals edit


<math>\int_a^x \!\!\!\int_a^s f(y)\,dy\,ds
= \int_a^x f(y)(x-y)\,dy</math>

Matrices and determinants edit


<math>\det(\mathsf{A}-\lambda\mathsf{I}) = 0</math>

Summation edit


<math>\sum_{i=0}^{n-1} i</math>



Differential equation edit


<math>u'' + p(x)u' + q(x)u=f(x),\quad x>a</math>

Complex numbers edit


<math>|\bar{z}| = |z|,
|(\bar{z})^n| = |z|^n,
\arg(z^n) = n \arg(z)</math>

Limits edit


<math>\lim_{z\rightarrow z_0} f(z)=f(z_0)</math>

Integral equation edit


<math>\phi_n(\kappa) =
\frac{1}{4\pi^2\kappa^2} \int_0^\infty
\frac{\sin(\kappa R)}{\kappa R}
\frac{\partial}{\partial R}
\left[R^2\frac{\partial D_n(R)}{\partial R}\right]\,dR</math>

Example edit


<math>\phi_n(\kappa) =

Continuation and cases edit


f(x) =
1 & -1 \le x < 0 \\
\frac{1}{2} & x = 0 \\
1 - x^2 & \text{otherwise}

Prefixed subscript edit


= \sum_{n=0}^\infty

Fraction and small fraction edit


<math>\frac{a}{b}\ \tfrac{a}{b}</math>

Area of a quadrilateral edit



Volume of a sphere-stand edit


<math>V=\frac16\pi h\left[3\left(r_1^2+r_2^2\right)+h^2\right]</math>

Multiple equations edit


u & = \tfrac{1}{\sqrt{2}}(x+y) \qquad & x &= \tfrac{1}{\sqrt{2}}(u+v) \\
v & = \tfrac{1}{\sqrt{2}}(x-y) \qquad & y &= \tfrac{1}{\sqrt{2}}(u-v)

See also edit

References edit

External links edit


  • Template:Citation. A paper introducing LaTeX — skip to page 49 for the math section. See page 63 for a complete reference list of symbols included in LaTeX and AMS-LaTeX.

Template:Help navigation