The free online HTML to Markdown converter turns a page or a snippet into clean Markdown, with GFM tables, fenced code and links kept intact. Because it runs on Turndown in the browser, the HTML is never uploaded.
Markup and Markdown work by OnlinePCApps since 2013
A block of HTML is heavy with tags a README, a wiki or a note cannot use. The free online HTML to Markdown converter strips it down to clean, portable Markdown.
A README, a GitHub wiki and a static site all speak Markdown, not HTML. Converting a page or a section moves the headings, links and code across as Markdown that renders cleanly wherever it lands.
A language model reads Markdown far better than raw HTML wrapped in scripts and styles. Converting first hands the model the words and the structure alone, which sharpens a summary or an answer.
An internal page or an unpublished draft should not pass through an outside server to be converted. Turndown runs in the browser, so the markup stays on the device while it becomes Markdown.
Paste HTML straight into the panel above, or drop an .html file from the device. A full page or a single snippet both work, and the markup is read where it sits.
Turn on GFM for tables, strikethrough and task lists. Set the heading, bullet and code style. The Markdown updates live as each option changes, so the result is easy to read before it is saved.
Copy the Markdown to the clipboard, or download it as a .md file. The source HTML stays untouched, so the same block can be converted again with different options at any time.
To convert HTML is to map each tag to its Markdown equivalent. The tool does that on the device with Turndown.
Each semantic tag becomes its Markdown equal, a heading to a hash, strong to a pair of stars, a link to bracket-and-parenthesis and an image to its own mark. The reader ends with clean text rather than a wall of tags.
A pre block with a code element becomes a fenced code block, and a class such as language-js carries the language onto the fence. Inline code stays wrapped in backticks, so a snippet reads and highlights the same as before.
With GitHub Flavored Markdown turned on, an HTML table becomes a pipe table, a strikethrough becomes a pair of tildes and a checkbox list becomes task items. Plain Markdown has no table syntax, so the GFM option covers it.
Scripts, inline styles and layout wrappers have no Markdown equal, so they are dropped rather than carried across as noise. What survives is the content and its structure, the part a README or a note actually needs.
To convert HTML to Markdown, to turn a page into an md file and to strip the tags all name the same task. A search for HTML to Markdown or Turndown online reaches this tool, and the source markup is left as it is.
A conversion maps each tag to its Markdown equal. These points decide how the result lands.
| The case | Result | What happens and why |
|---|---|---|
| Headings and text | 1:1 | An h1 to h6 becomes a hash heading, and bold or italic becomes stars or underscores. |
| Links and images | kept intact | A link keeps its text and href, and an image keeps its alt text and src. |
| Code blocks | fenced | A pre and code block becomes a fenced block, with the language from a class carried on. |
| Lists | nested | Ordered and unordered lists convert, and nesting is kept by indentation. |
| Tables | GFM only | A table becomes a pipe table when GitHub Flavored Markdown is turned on. |
| Strikethrough, tasks | GFM only | Strikethrough and checkbox task lists also need the GFM option switched on. |
| Scripts and styles | dropped | Scripts, inline styles and layout wrappers have no Markdown equal and are removed. |
| No Markdown equal | kept as HTML | A tag Markdown cannot express can be left inline as raw HTML instead of lost. |
| Where it runs | on the device | The markup is converted in the browser, so nothing is uploaded. |
From HTML Tags to Markdown
HTML and Markdown describe the same content at two very different levels of detail. HTML has more than a hundred element types and can express layout, colour and behaviour through tags, classes and styles. Markdown has around fifteen constructs and covers headings, emphasis, links, images, lists, blockquotes and code. A converter maps the semantic tags onto their Markdown equal, so an h2 becomes a hash heading, a strong becomes a pair of stars, a link keeps its text and address and a code block keeps its fence and language. This tool runs on Turndown, the most widely used library for the job at more than three million installs a week, which follows the CommonMark rules, with the GFM plugin adding tables, strikethrough and task lists. The honest point is that the mapping is lossy on purpose. A styled div, an inline colour, a form or a script has no Markdown syntax, so it is dropped or left as inline HTML. The result favours a clean, portable file for a README, a wiki or a note over a pixel-perfect copy of the page, which is almost always the point of moving to Markdown. The references below define the formats.
Both turn HTML into Markdown. The trade is real, and the source is often internal.
| Point of comparison | This tool Built in the browser On the device | pandoc A local command-line tool |
|---|---|---|
| Where the HTML goes | Stays on the device | Stays on the machine |
| Price and caps | Free with no file cap | Free tier often capped |
| Many document formats | HTML to Markdown only | DOCX, LaTeX and dozens more |
| A book-length document | Held in tab memory | Streams from disk |
| Script a build pipeline | A paste at a time here | One command in a script |
pandoc wins where the job spans many document formats, a book-length file or a scripted build. This page wins for the everyday case, a page or a snippet turned to clean Markdown in a couple of clicks. Both keep the source local, and this one needs nothing installed.
The markup is parsed and rewritten inside the browser by Turndown, so the work is client-side and the HTML stays on the machine that opened it. No copy is passed to a server for the job.
HTML can hold an internal page, an unpublished draft or a client email never meant to sit on an outside server. A hosted converter has to upload the markup to process it, more exposure than a format change is worth.
Each of these turns HTML into Markdown. They differ in effort, in cost and in where the source goes.
pandoc, Turndown in code and a hand clean all convert HTML well, yet each needs an install, a project or slow retyping. This page runs the same Turndown engine on a paste or a file, kept local with nothing to set up.
A little context sets what to expect.
Markdown covers headings, links, lists, quotes, images and code. A styled layout, a colour or a form has no Markdown syntax, so it is dropped or kept as inline HTML. The output is clean content rather than a copy of the page.
Plain Markdown has no table syntax, and no strikethrough or task lists. Switching on GitHub Flavored Markdown adds all three, so a table in the HTML comes out as a pipe table rather than raw tags.
Because the work is client-side, the tool reads HTML that is pasted in or dropped as a file. It does not fetch a live web address, since that would need a server. Copy the page source, or save it as an .html file first.
The browser converts a page or a snippet in memory, which fits an everyday paste. A folder of HTML files, or a whole site moving to a Markdown docs system, belongs on the desktop edition, which reads from disk and writes each file in one pass.
Point it at a folder of HTML files and each one is written to its own Markdown file in a single run, saved beside the source.
A book-length page or a long export is streamed from disk, past what a browser tab keeps in memory at once.
Saved presets set the GFM options and the heading and code style once, ready to apply across a repeated migration.
Free and online, with no sign-up and no upload. Turn HTML into clean Markdown in the browser, all on the device.