← Workers Ai / workers-ai / features / markdown-conversion / usage

REST API

You can also use the Markdown Conversion REST API to convert your documents into Markdown.

Prerequisite: Get Workers AI API token

To use the Markdown Conversion service via the REST API, you need an API token with permissions for the Workers AI REST API. Refer to Get started with the Workers AI REST API for instructions on obtaining an API token with the correct permissions.

Transform

This endpoint lets you convert any file given to us into markdown.

curl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/tomarkdown \
  -X POST \
  -H 'Authorization: Bearer {API_TOKEN}' \
  -F "[email protected]" \
  -F "[email protected]" \
  -F 'conversionOptions={ ... }'

Parameters

files File[] required

The files you want to convert.

conversionOptions ConversionOptions optional

Options that allow you to control how your files are converted. Refer to Conversion Options for further details.

Response

{
	"success": true,
	"result": [
		{
			"id": "...",
			"name": "good.html",
			"mimeType": "text/html",
			"format": "markdown",
			"tokens": 49,
			"data": "# Image Embedded with a Data URI\n\nThis _image_ is directly encoded in the HTML:\n\n\n\nAn image description\n\n \n\nIt's a tiny 5x5 pixel PNG, scaled up to 50x50px.\n\n"
		},
		{
			"id": "...",
			"name": "bad.pdf",
			"mimeType": "application/pdf",
			"format": "error",
			"error": "Some error that prevented this image from being converted"
		}
	]
}

Supported

This endpoint lets you programmatically retrieve the full set of rich formats that are supported for conversion.

curl https://api.cloudflare.com/client/v4/accounts/{ACCOUNT_ID}/ai/tomarkdown/supported \
  -H 'Authorization: Bearer {API_TOKEN}'

Response

{
  "success": true,
  "result": [
    {
      "extension": ".html",
      "mimeType": "text/html"
    },
    {
      "extension": ".pdf",
      "mimeType": "application/pdf"
    },
    ...
  ]
}