Skip to content

Chart composition

Creating advanced custom charts.


The @mui/x-charts follows an architecture based on context providers. The overall idea is to pass your series and axes definitions to a single component: the <ChartContainer />. This component transforms the data and makes it available to its children.

Based on the data provided by the container, you can render some graphical elements with provided subcomponents, such as <LinePlot /> or <ChartsYAxis />. Or you can create your own components.

Container options


There are two types of Chart containers available: <ChartContainer /> and <ResponsiveChartContainer />. As the names suggest, the only difference between them is responsiveness.

The first container requires you to provide width and height props. In contrast, <ResponsiveChartContainer /> automatically adjusts its dimensions to fit the available space defined by the parent element.

The following demo lets you switch between a chart using <ChartContainer /> with width set to 500 and height set to 300, and a chart using <ResponsiveChartContainer />, so you can see how they differ.



The chart container gets all props that are not specific to a single graphical element. This includes:

  • The xAxis and yAxis props—find more information in the Axis doc
  • The colors prop as defined in the color palette page
  • The series and dataset props


The series prop is an array of series definitions. You can find an explanation about each specific series type in their respective docs page: Line, Bar, Pie, and Scatter.

When using a single Charts component, the library can guess which kind of series you are defining. For example, the Bar Chart component assumes that series will be of type 'bar'.

With composition, the chart container isn't able to guess the series type, so you must explicitly define it.

<BarChart series={[{
    data: [1, 2, 3] // No need to specify it is a bar series
}]} />

    { data: [1, 2, 3], type: 'bar' }, // This series is for the bar chart
    { data: [3, 2, 1], type: 'line' } // This series is for the line chart
  <BarPlot /> {/* Will only display series with type: 'bar' */}
  <LinePlot /> {/* Will only display series with type: 'line' */}

Those series can use the dataset prop the same way that a single-component chart does—see Using a dataset in the Bar Chart documentation for more details.

In the next demo, the chart is made by composing the <BarPlot /> and <LinePlot /> components. By modifying the series type property, you can switch between rendering a line and a bar.

    { type, data: [1, 2, 3, 2, 1] },
    { type, data: [4, 3, 1, 3, 4] },
  <BarPlot />
  <LinePlot />
  <ChartsXAxis label="X axis" position="bottom" axisId="x-axis-id" />



To display data, you have components named <XxxPlot /> such as <LinePlot />, <AreaPlot />, <MarkPlot />, <BarPlot />, etc.


To ensure chart elements stay confined to the designated drawing area, use the ChartsClipPath component. This component defines a rectangular clip path that acts as a boundary.

  1. Define the Clip Path: Use <ChartsClipPath id={clipPathId} /> to establish the clip path for the drawing area. clipPathId must be a unique identifier.
  2. Wrap the Chart: Enclose the chart elements you want to clip within a <g> element. Set the clipPath attribute to url(#${clipPathId}) to reference the previously defined clip path. Example: <g clipPath={`url(#${clipPathId})`}>
  <g clipPath={`url(#${clipPathId})`}>
    // The plotting to clip in the drawing area.
    <ScatterPlot />
    <LinePlot />
  <ChartsClipPath id={clipPathId} /> // Defines the clip path of the drawing area.

The following demo allows you to toggle clipping for scatter and line plots. Observe how line markers extend beyond the clip area, rendering on top of the axes.


To add axes, you can use <ChartsXAxis /> and <ChartsYAxis /> as defined in the axis page.

It takes an axisId prop that indicates which axis, defined in the container, should be rendered. If axisId is not provided it will pick the first one.


To add a grid, you can use the <ChartsGrid /> component.

See Axis—Grid documentation for more information.

Additional information

To add a legend to your chart, you can use <ChartsLegend />.

Most of the props are explained in the legend page. The demos use the slotProps.legend object, but with composition, you can pass props directly to <ChartsLegend />.

// With single component chart
    legend: {
      direction: 'row',

// With composition
  <ChartsLegend direction="row" />


You can also add interactive elements such as <ChartsAxisHighlight /> and <ChartsTooltip />.


See the documentation below for a complete reference to all of the props and classes available to the components mentioned here.