CSS values and units

In CSS specifications and on the property pages here on MDN you will be able to spot value types as they will be surrounded by angle brackets, such as <color> or <length>. When you see the value type <color> as valid for a particular property, that means you can use any valid color as a value for that property, as listed on the <color> reference page.

In the following example, we have set the color of our heading using a keyword, and the background using the rgb() function:

h1 {
  color: black;
  background-color: rgb(197 93 161);
}

A value type in CSS is a way to define a collection of allowable values. This means that if you see <color> as valid you don't need to wonder which of the different types of color value can be used — keywords, hex values, rgb() functions, etc. You can use any available <color> values, assuming they are supported by your browser. The page on MDN for each value will give you information about browser support. For example, if you look at the page for <color> you will see that the browser compatibility section lists different types of color values and support for them.

Let's have a look at some of the types of values and units you may frequently encounter, with examples so that you can try out different possible values.

There are various numeric value types that you might find yourself using in CSS. The following are all classed as numeric:

The numeric type you will come across most frequently is <length>. For example, 10px (pixels) or 30em. There are two types of lengths used in CSS — relative and absolute. It's important to know the difference in order to understand how big things will become.

Absolute length units

The following are all absolute length units — they are not relative to anything else, and are generally considered to always be the same size.

Most of these units are more useful when used for print, rather than screen output. For example, we don't typically use cm (centimeters) on screen. The only value that you will commonly use is px (pixels).

Relative length units

Relative length units are relative to something else. For example:

• em is relative to the font size of this element, or the font size of the parent element when used for font-size. rem is relative to the font size of the root element.
• vh and vw are relative to the viewport's height and width, respectively.

The benefit of using relative units is that with some careful planning you can make it so the size of text or other elements scales relative to everything else on the page. For a complete list of the relative units available, see the reference page for the <length> type.

In this section we'll explore some of the most common relative units.

Exploring an example

In the example below, you can see how some relative and absolute length units behave. The first box has a width set in pixels. As an absolute unit, this width will remain the same no matter what else changes.

The second box has a width set in vw (viewport width) units. This value is relative to the viewport width, and so 10vw is 10 percent of the width of the viewport. If you change the width of your browser window, the size of the box should change. However this example is embedded into the page using an <iframe>, so this won't work. To see this in action you'll have to try the example after opening it in its own browser tab.

The third box uses em units. These are relative to the element's font size. I've set a font size of 1em on the containing <div>, which has a class of .wrapper. Change this value to 1.5em and you will see that the font size of all the elements increases, but only the last item will get wider, as its width is relative to that font size.

After following the instructions above, try playing with the values in other ways, to see what you get.

ems and rems

em and rem are the two relative lengths you are likely to encounter most frequently when sizing anything from boxes to text. It's worth understanding how these work, and the differences between them, especially when you start getting on to more complex subjects like styling text or CSS layout. The below example provides a demonstration.

The HTML illustrated below is a set of nested lists — we have two lists in total and both examples have the same HTML. The only difference is that the first has a class of ems and the second a class of rems.

To start with, we set 16px as the font size on the <html> element.

To recap, the em unit means "my parent element's font-size" if used for font-size (and "my own font-size" when used for anything else). The <li> elements inside the <ul> with a class of ems take their sizing from their parent. So each successive level of nesting gets progressively larger, as each has its font size set to 1.3em — 1.3 times its parent element's font size.

To recap, the rem unit means "The root element's font-size" (rem stands for "root em"). The <li> elements inside the <ul> with a class of rems take their sizing from the root element (<html>). This means that each successive level of nesting does not keep getting larger.

However, if you change the <html> element's font-size in the CSS you will see that everything else changes relative to it — both rem- and em-sized text.

Line height units

lh and rlh are relative lengths units similar to em and rem. The difference between lh and rlh is that the first one is relative to the line height of the element itself, while the second one is relative to the line height of the root element, usually <html>.

Using these units, we can precisely align box decoration to the text. In this example, we use lh unit to create notepad-like lines using repeating-linear-gradient(). It doesn't matter what's the line height of the text, the lines will always start in the right place.

body {
  margin: 0;
  display: grid;
  grid-template-columns: 1fr 1fr;
  padding: 24px;
  gap: 24px;
  background-color: floralwhite;
  font-family: sans-serif;
}

@supports not (height: 1lh) {
  body::before {
    grid-column: 1 / -1;
    padding: 8px;
    border-radius: 4px;
    background-color: tomato;
    color: white;
    content: "You browser doesn't support lh unit just yet";
  }
}

p {
  margin: 0;
  background-image: repeating-linear-gradient(
    to top,
    lightskyblue 0 2px,
    transparent 2px 1lh
  );
}

<p style="line-height: 2em">
  Summer is a time for adventure, and this year was no exception. I had many
  exciting experiences, but two of my favorites were my trip to the beach and my
  week at summer camp.
</p>

<p style="line-height: 4em">
  At the beach, I spent my days swimming, collecting shells, and building
  sandcastles. I also went on a boat ride and saw dolphins swimming alongside
  us.
</p>

In a lot of cases, a percentage is treated in the same way as a length. The thing with percentages is that they are always set relative to some other value. For example, if you set an element's font-size as a percentage, it will be a percentage of the font-size of the element's parent. If you use a percentage for a width value, it will be a percentage of the width of the parent.

In the below example the two percentage-sized boxes and the two pixel-sized boxes have the same class names. The sets are 40% and 200px wide respectively.

The difference is that the second set of two boxes is inside a wrapper that is 400 pixels wide. The second 200px wide box is the same width as the first one, but the second 40% box is now 40% of 400px — a lot narrower than the first one!

Try changing the width of the wrapper or the percentage value to see how this works.

The next example has font sizes set in percentages. Each <li> has a font-size of 80%; therefore, the nested list items become progressively smaller as they inherit their sizing from their parent.

Note that, while many value types accept a length or a percentage, there are some that only accept length. You can see which values are accepted on the MDN property reference pages. If the allowed value includes <length-percentage> then you can use a length or a percentage. If the allowed value only includes <length>, it is not possible to use a percentage.

Some value types accept numbers, without any unit added to them. An example of a property which accepts a unitless number is the opacity property, which controls the opacity of an element (how transparent it is). This property accepts a number between 0 (fully transparent) and 1 (fully opaque).

In the below example, try changing the value of opacity to various decimal values between 0 and 1 and see how the box and its contents become more or less opaque.

Color values can be used in many places in CSS, whether you are specifying the color of text, backgrounds, borders, and lots more. There are many ways to set color in CSS, allowing you to control plenty of exciting properties.

The standard color system available in modern computers supports 24-bit colors, which allows displaying about 16.7 million distinct colors via a combination of different red, green, and blue channels with 256 different values per channel (256 x 256 x 256 = 16,777,216).

In this section, we'll first look at the most commonly seen ways of specifying colors: using keywords, hexadecimal, and rgb() values. We'll also take a quick look at additional color functions, enabling you to recognize them when you see them or experiment with different ways of applying color.

You will likely decide on a color palette and then use those colors — and your favorite way of specifying color — throughout your project. You can mix and match color models, but it's usually best if your entire project uses the same method of declaring colors for consistency!

You will see the color keywords (or 'named colors') used in many MDN code examples. As the <named-color>s data type contains a very finite number of color values, these are not commonly used on production websites. As the keyword represents the color as a human-readable text value, named colors are used in code examples to clearly tell the user what color is expected so the learner can focus on the content being taught.

Try playing with different color values in the live examples below, to get more of an idea how they work.

The next type of color value you are likely to encounter is hexadecimal codes. Hexadecimal uses 16 characters from 0-9 and a-f, so the entire range is 0123456789abcdef. Each hex color value consists of a hash/pound symbol (#) followed by three or six hexadecimal characters (#fcc or #ffc0cb, for example), with an optional one or two hexadecimal characters representing the alpha-transparency of the previous three or six character color values.

When using hexadecimal to describe RGB values, each pair of hexadecimal characters is a decimal number representing one of the channels — red, green and blue — and allows us to specify any of the 256 available values for each (16 x 16 = 256). These values are less intuitive than keywords for defining colors, but they are a lot more versatile because you can represent any RGB color with them.

Again, try changing the values to see how the colors vary.

To create RGB values directly, the rgb() function takes three parameters representing red, green, and blue channel values of the colors, with an optional fourth value separated by a slash ('/') representing opacity, in much the same way as hex values. The difference with RGB is that each channel is represented not by two hex digits, but by a decimal number between 0 and 255 or a percent between 0% and 100% inclusive (but not a mixture of the two).

Let's rewrite our last example to use RGB colors:

You can pass a fourth parameter to rgb(), which represents the alpha channel of the color, which controls opacity. If you set this value to 0 it will make the color fully transparent, whereas 1 will make it fully opaque. Values in between give you different levels of transparency.

In the example below, we have added a background image to the containing block of our colored boxes. We have then set the boxes to have different opacity values — notice how the background shows through more when the alpha channel value is smaller.

In this example, try changing the alpha channel values to see how it affects the color output.

The sRGB color space defines colors in the red (r), green (g), and blue (b) color space.

If you want to go beyond keywords, hexadecimal, and rgb() for colors, you might want to try using <hue>. Hue is the property that allows us to tell the difference or similarity between colors like red, orange, yellow, green, blue, etc. The key concept is that you can specify a hue in an <angle> because most of the color models describe hues using a color wheel.

There are several color functions that include a <hue> component, including hsl(), hwb(), and lch(). Other color functions, like lab(), define colors based on what humans can see.

If you want to find out more about these functions and color spaces, see the Applying color to HTML elements using CSS guide, the <color> reference that lists all the different ways you can use colors in CSS, and the CSS color module that provides an overview of all the color types in CSS and the properties that use color values.

A great starting point for using hues in CSS is the hwb() function which specifies an srgb() color. The three parts are:

• Hue: The base shade of the color. This takes a <hue> value between 0 and 360, representing the angles around a color wheel.
• Whiteness: How white is the color? This takes a value from 0% (no whiteness) to 100% (full whiteness).
• Blackness: How black is the color? This takes a value from 0% (no blackness) to 100% (full blackness).

Similar to the hwb() function is the hsl() function which also specifies an srgb() color. HSL uses Hue, in addition to Saturation and Lightness:

• Hue
• Saturation: How saturated is the color? This takes a value from 0–100%, where 0 is no color (it will appear as a shade of grey), and 100% is full color saturation.
• Lightness: How light or bright is the color? This takes a value from 0–100%, where 0 is no light (it will appear completely black) and 100% is full light (it will appear completely white).

The hsl() color value also has an optional fourth value, separated from the color with a slash (/), representing the alpha transparency.

Let's update the RGB example to use HSL colors instead:

Just like with rgb() you can pass an alpha parameter to hsl() to specify opacity:

The <image> value type is used wherever an image is a valid value. This can be an actual image file pointed to via a url() function, or a gradient.

In the example below, we have demonstrated an image and a gradient in use as a value for the CSS background-image property.

The <position> value type represents a set of 2D coordinates, used to position an item such as a background image (via background-position). It can take keywords such as top, left, bottom, right, and center to align items with specific bounds of a 2D box, along with lengths, which represent offsets from the top and left-hand edges of the box.

A typical position value consists of two values — the first sets the position horizontally, the second vertically. If you only specify values for one axis the other will default to center.

In the following example we have positioned a background image 40px from the top and to the right of the container using a keyword.

Play around with these values to see how you can push the image around.

Throughout the examples above, we've seen places where keywords are used as a value (for example <color> keywords like red, black, rebeccapurple, and goldenrod). These keywords are more accurately described as identifiers, a special value that CSS understands. As such they are not quoted — they are not treated as strings.

There are places where you use strings in CSS. For example, when specifying generated content. In this case, the value is quoted to demonstrate that it is a string. In the example below, we use unquoted color keywords along with a quoted generated content string.

In programming, a function is a piece of code that does a specific task. Functions are useful because you can write code once, then reuse it many times instead of writing the same logic over and over. Most programming languages not only support functions but also come with convenient built-in functions for common tasks so you don't have to write them yourself from scratch.

CSS also has functions, which work in a similar way to functions in other languages. In fact, we've already seen CSS functions in the Color section above with rgb() and hsl() functions.

Aside from applying colors, you can use functions in CSS to do a lot of other things. For example Transform functions are a common way to move, rotate, and scale elements on a page. You might see translate() for moving something horizontally or vertically, rotate() to rotate something, or scale() to make something bigger or smaller.

When you are creating styles for a project, you will probably start off with numbers like 300px for lengths or 200ms for durations. If you want to have these values change based on other values, you will need to do some math. You could calculate the percentage of a value or add a number to another number, then update your CSS with the result.

CSS has support for Math functions, which allow us to perform calculations instead of relying on static values or doing the math in JavaScript. One of the most common math functions is calc() which lets you do operations like addition, subtraction, multiplication, and division.

For example, let's say we want to set the width of an element to be 20% of its parent container plus 100px. We can't specify this width with a static value — if the parent uses a percentage width (or a relative unit like em or rem) then it will vary depending on the context it is used in, and other factors such as the user's device or browser window width. However, we can use calc() to set the width of the element to be 20% of its parent container plus 100px. The 20% is based on the width of the parent container (.wrapper) and if that width changes, the calculation will change too:

There are many other math functions that you can use in CSS, such as min(), max(), and clamp(); respectively these let you pick the smallest, largest, or middle value from a set of values. You can also use Trigonometric functions like sin(), cos(), and tan() to calculate angles for rotating elements around a point, or choose colors that take a hue angle as a parameter. Exponential functions might also be used for animations and transitions, when you require very specific control over how something moves and looks.

Knowing about CSS functions is useful so you recognize them when you see them. You should start experimenting with them in your projects — they will help you avoid writing custom or repetitive code to achieve results that you can get with regular CSS.

You've reached the end of this article, but can you remember the most important information? You can find some further tests to verify that you've retained this information before you move on — see Test your skills: Values and units.

This has been a quick run-through of the most common types of values and units you might encounter. You can have a look at all of the different types on the CSS Values and units reference page — you will encounter many of these in use as you work through these lessons.

The key thing to remember is that each property has a defined list of allowed value types, and each value type has a definition explaining what the values are. You can then look up the specifics here on MDN. For example, understanding that <image> also allows you to create a color gradient is useful but perhaps non-obvious knowledge to have!

In the next article, we'll take a look at how items are sized in CSS.

• Previous
• Overview: CSS building blocks
• Next