Markdown

Parse and render Markdown with Bun's built-in Markdown API, supporting GFM extensions and custom rendering callbacks

Unstable API — This API is under active development and may change in future versions of Bun.

Bun includes a fast, built-in Markdown parser written in Zig. It supports GitHub Flavored Markdown (GFM) extensions and provides three APIs:

Bun.markdown.html() — render Markdown to an HTML string
Bun.markdown.render() — render Markdown with custom callbacks for each element
Bun.markdown.react() — render Markdown to React JSX elements

`Bun.markdown.html()`

Convert a Markdown string to HTML.

const html = Bun.markdown.html("# Hello **world**");
// "<h1>Hello <strong>world</strong></h1>\n"

GFM extensions like tables, strikethrough, and task lists are enabled by default:

const html = Bun.markdown.html(`
| Feature      | Status |
|-------------|--------|
| Tables       | ~~done~~ |
| Strikethrough| ~~done~~ |
| Task lists   | done |
`);

Options

Pass an options object as the second argument to configure the parser:

const html = Bun.markdown.html("some markdown", {
  tables: true, // GFM tables (default: true)
  strikethrough: true, // GFM strikethrough (default: true)
  tasklists: true, // GFM task lists (default: true)
  tagFilter: true, // GFM tag filter for disallowed HTML tags
  autolinks: true, // Autolink URLs, emails, and www. links
});

All available options:

Option	Default	Description
`tables`	`false`	GFM tables
`strikethrough`	`false`	GFM strikethrough (`~~text~~`)
`tasklists`	`false`	GFM task lists (`- [x] item`)
`autolinks`	`false`	Enable autolinks — see Autolinks
`headings`	`false`	Heading IDs and autolinks — see Heading IDs
`hardSoftBreaks`	`false`	Treat soft line breaks as hard breaks
`wikiLinks`	`false`	Enable `[[wiki links]]`
`underline`	`false`	`__text__` renders as `<u>` instead of `<strong>`
`latexMath`	`false`	Enable $inline$ and `$$display$$` math
`collapseWhitespace`	`false`	Collapse whitespace in text
`permissiveAtxHeaders`	`false`	ATX headers without space after `#`
`noIndentedCodeBlocks`	`false`	Disable indented code blocks
`noHtmlBlocks`	`false`	Disable HTML blocks
`noHtmlSpans`	`false`	Disable inline HTML
`tagFilter`	`false`	GFM tag filter for disallowed HTML tags

Autolinks

Pass true to enable all autolink types, or an object for granular control:

// Enable all autolinks (URL, WWW, email)
Bun.markdown.html("Visit www.example.com", { autolinks: true });

// Enable only specific types
Bun.markdown.html("Visit www.example.com", {
  autolinks: { url: true, www: true },
});

Heading IDs

Pass true to enable both heading IDs and autolink headings, or an object for granular control:

// Enable heading IDs and autolink headings
Bun.markdown.html("## Hello World", { headings: true });
// '<h2 id="hello-world"><a href="#hello-world">Hello World</a></h2>\n'

// Enable only heading IDs (no autolink)
Bun.markdown.html("## Hello World", { headings: { ids: true } });
// '<h2 id="hello-world">Hello World</h2>\n'

Parse Markdown and render it using custom JavaScript callbacks. This gives you full control over the output format — you can generate HTML with custom classes, React elements, ANSI terminal output, or any other string format.

const result = Bun.markdown.render("# Hello **world**", {
  heading: (children, { level }) => `<h${level} class="title">${children}</h${level}>`,
  strong: children => `<b>${children}</b>`,
  paragraph: children => `<p>${children}</p>`,
});
// '<h1 class="title">Hello <b>world</b></h1>'

Callback signature

Each callback receives:

children — the accumulated content of the element as a string
meta (optional) — an object with element-specific metadata

Return a string to replace the element's rendering. Return null or undefined to omit the element from the output entirely. If no callback is registered for an element, its children pass through unchanged.

Block callbacks

Callback	Meta	Description
`heading`	`{ level: number, id?: string }`	Heading level 1–6. `id` is set when `headings: { ids: true }` is enabled
`paragraph`	—	Paragraph block
`blockquote`	—	Blockquote block
`code`	`{ language?: string }`	Fenced or indented code block. `language` is the info-string when specified on the fence
`list`	`{ ordered: boolean, start?: number }`	Ordered or unordered list. `start` is the start number for ordered lists
`listItem`	`{ checked?: boolean }`	List item. `checked` is set for task list items (`- [x]` / `- [ ]`)
`hr`	—	Horizontal rule
`table`	—	Table block
`thead`	—	Table head
`tbody`	—	Table body
`tr`	—	Table row
`th`	`{ align?: "left" \| "center" \| "right" }`	Table header cell. `align` is set when alignment is specified
`td`	`{ align?: "left" \| "center" \| "right" }`	Table data cell. `align` is set when alignment is specified
`html`	—	Raw HTML content

Inline callbacks

Callback	Meta	Description
`strong`	—	Strong emphasis (`text`)
`emphasis`	—	Emphasis (`text`)
`link`	`{ href: string, title?: string }`	Link
`image`	`{ src: string, title?: string }`	Image
`codespan`	—	Inline code (`code`)
`strikethrough`	—	Strikethrough (`~~text~~`)
`text`	—	Plain text content

Examples

Custom HTML with classes

const html = Bun.markdown.render("# Title\n\nHello **world**", {
  heading: (children, { level }) => `<h${level} class="heading heading-${level}">${children}</h${level}>`,
  paragraph: children => `<p class="body">${children}</p>`,
  strong: children => `<strong class="bold">${children}</strong>`,
});

Stripping all formatting

const plaintext = Bun.markdown.render("# Hello **world**", {
  heading: children => children,
  paragraph: children => children,
  strong: children => children,
  emphasis: children => children,
  link: children => children,
  image: () => "",
  code: children => children,
  codespan: children => children,
});
// "Hello world"

Omitting elements

Return null or undefined to remove an element from the output:

const result = Bun.markdown.render("# Title\n\n![logo](img.png)\n\nHello", {
  image: () => null, // Remove all images
  heading: children => children,
  paragraph: children => children + "\n",
});
// "Title\nHello\n"

ANSI terminal output

const ansi = Bun.markdown.render("# Hello\n\nThis is **bold** and *italic*", {
  heading: (children, { level }) => `\x1b[1;4m${children}\x1b[0m\n`,
  paragraph: children => children + "\n",
  strong: children => `\x1b[1m${children}\x1b[22m`,
  emphasis: children => `\x1b[3m${children}\x1b[23m`,
});

Code block syntax highlighting

const result = Bun.markdown.render("```js\nconsole.log('hi')\n```", {
  code: (children, meta) => {
    const lang = meta?.language ?? "";
    return `<pre><code class="language-${lang}">${children}</code></pre>`;
  },
});

Parser options

Parser options are passed as a separate third argument:

const result = Bun.markdown.render(
  "Visit www.example.com",
  {
    link: (children, { href }) => `[${children}](${href})`,
    paragraph: children => children,
  },
  { autolinks: true },
);

`Bun.markdown.react()`

Render Markdown directly to React elements. Returns a <Fragment> that you can use as a component return value.

function Markdown({ text }: { text: string }) {
  return Bun.markdown.react(text);
}

Server-side rendering

Works with renderToString() and React Server Components:

import { renderToString } from "react-dom/server";

const html = renderToString(Bun.markdown.react("# Hello **world**"));
// "<h1>Hello <strong>world</strong></h1>"

Component overrides

Replace any HTML element with a custom React component by passing it in the second argument, keyed by tag name:

function Code({ language, children }) {
  return (
    <pre data-language={language}>
      <code>{children}</code>
    </pre>
  );
}

function Link({ href, title, children }) {
  return (
    <a href={href} title={title} target="_blank" rel="noopener noreferrer">
      {children}
    </a>
  );
}

function Heading({ id, children }) {
  return (
    <h2 id={id}>
      <a href={`#${id}`}>{children}</a>
    </h2>
  );
}

const el = Bun.markdown.react(
  content,
  {
    pre: Code,
    a: Link,
    h2: Heading,
  },
  { headings: { ids: true } },
);

Available overrides

Every HTML tag produced by the parser can be overridden:

Option	Props	Description
`h1`–`h6`	`{ id?, children }`	Headings. `id` is set when `headings: { ids: true }` is enabled
`p`	`{ children }`	Paragraph
`blockquote`	`{ children }`	Blockquote
`pre`	`{ language?, children }`	Code block. `language` is the info string (e.g. `"js"`)
`hr`	`{}`	Horizontal rule (no children)
`ul`	`{ children }`	Unordered list
`ol`	`{ start, children }`	Ordered list. `start` is the first item number
`li`	`{ checked?, children }`	List item. `checked` is set for task list items
`table`	`{ children }`	Table
`thead`	`{ children }`	Table head
`tbody`	`{ children }`	Table body
`tr`	`{ children }`	Table row
`th`	`{ align?, children }`	Table header cell
`td`	`{ align?, children }`	Table data cell
`em`	`{ children }`	Emphasis (`text`)
`strong`	`{ children }`	Strong (`text`)
`a`	`{ href, title?, children }`	Link
`img`	`{ src, alt?, title? }`	Image (no children)
`code`	`{ children }`	Inline code
`del`	`{ children }`	Strikethrough (`~~text~~`)
`br`	`{}`	Hard line break (no children)

React 18 and older

By default, elements use Symbol.for('react.transitional.element') as the $$typeof symbol. For React 18 and older, pass reactVersion: 18 in the options (third argument):

function Markdown({ text }: { text: string }) {
  return Bun.markdown.react(text, undefined, { reactVersion: 18 });
}

Parser options

All parser options are passed as the third argument:

const el = Bun.markdown.react("## Hello World", undefined, {
  headings: { ids: true },
  autolinks: true,
});

`Bun.markdown.html()`

Options

Autolinks

Heading IDs

`Bun.markdown.render()`

Callback signature

Block callbacks

Inline callbacks

Examples

Custom HTML with classes

Stripping all formatting

Omitting elements

ANSI terminal output

Code block syntax highlighting

Parser options

`Bun.markdown.react()`

Server-side rendering

Component overrides

Available overrides

React 18 and older

Parser options

On this page