Response and headers

Markdown (default)

With response_format=markdown you receive the converted document as a raw body:

HTTP/1.1 200 OK
content-type: text/markdown; charset=utf-8
content-disposition: attachment; filename="report.md"

JSON

With response_format=json you receive an object with the Markdown and its metadata:

{
  "filename": "report.pdf",
  "input_format": "pdf",
  "markdown": "# Informe...\n\n...",
  "size_bytes": 184320,
  "timings": {
    "convert_ms": 142,
    "clean_ms": 3,
    "total_worker_ms": 145,
    "upload_ms": 20,
    "total_request_ms": 171
  },
  "markdown_bytes": 5120,
  "token_saved_estimate": 44800
}

string

Document converted to Markdown.

string

Detected or forced format of the input.

string

integer

Size of the processed document in bytes.

object

Times in milliseconds: convert_ms, clean_ms, total_worker_ms, upload_ms, total_request_ms.

integer

Size of the output Markdown in bytes.

integer

Estimation of tokens saved versus entering the raw document into the prompt.

Headers

Every response includes observability headers, useful for logging and metrics in your client.

Header	Meaning
`X-Conversion-Engine`	Used conversion engine.
`X-Input-Format`	Detected input format.
`X-Input-Bytes`	Bytes of the processed document.
`X-Wire-Bytes`	Bytes received over network (`/convert/raw` with compression only).
`X-Markdown-Bytes`	Output Markdown size.
`X-Token-Saved-Estimate`	Estimated tokens saved.
`X-Conversion-Ms`	Conversion time.
`X-Total-Request-Ms`	Total time of the request.
`ETag`	Hash of the content + parameters. Use it for cache.
`Cache-Control`	`public, max-age=..., immutable`.

See Cache and ETag.

​Response and headers

​Markdown (default)

​JSON

​Headers

Response and headers

Markdown (default)

JSON

Headers