Nepali PDF API

Overview

Base URL	https://nepalipdf.com
Upload	multipart/form-data, the PDF in a field named file
Limits	20 MB and 50 pages per PDF; see limits
Errors	JSON: {"error": "…"} with a 4xx or 5xx status
Privacy	Nothing is kept: files are deleted as soon as the response is sent

Requests are processed one at a time; if the service is busy, your request waits its turn. Allow a generous timeout (a few minutes) for long documents.

PDF to Word

POST /api/v1/pdf-to-word

Returns the Word document (.docx) as the response body.

Field	Required	Description
file	yes	The PDF.
ascii_digits	no	true to write Nepali digits (०१२३) as 0123. Default false.
tables	no	How tables are found: auto (default, from ruling lines), text (from text alignment, for tables without lines) or none.

curl -X POST https://nepalipdf.com/api/v1/pdf-to-word \
  -F "file=@notice.pdf" \
  -o notice.docx

PDF to text or Markdown

POST /api/v1/pdf-to-text

Returns UTF-8 text as the response body: text/plain or text/markdown.

Field	Required	Description
file	yes	The PDF.
format	no	text (default) or markdown. Plain text separates table columns with tabs; Markdown keeps headings and writes tables as pipe tables.
ascii_digits	no	As above.
tables	no	As above.

curl -X POST https://nepalipdf.com/api/v1/pdf-to-text \
  -F "file=@notice.pdf" \
  -F "format=markdown" \
  -o notice.md

Response headers

Both endpoints add a short summary of the conversion:

X-Pages	Number of pages in the PDF.
X-Flags	How many things are worth checking by eye. 0 means every character was identified exactly; anything else means some text may be wrong, for example because the PDF uses a font the service has not seen before.
X-Elapsed-Seconds	Time spent converting.

Python example

import requests

with open("notice.pdf", "rb") as pdf:
    response = requests.post(
        "https://nepalipdf.com/api/v1/pdf-to-text",
        files={"file": pdf},
        data={"format": "text"},
        timeout=300,
    )
response.raise_for_status()
print(response.text)
print("flags:", response.headers["X-Flags"])

Errors

Status	Meaning
400	Missing file or an invalid field value.
413	The file is larger than 20 MB, or has more than 50 pages.
415	The file is not a PDF.
422	The PDF took longer than 120 seconds to convert and was stopped.
429	Too many conversions from your address, or your previous conversion is still in progress. Wait for the number of seconds in the Retry-After header.
500	The PDF could not be converted.
503	The service is busy. Try again after the Retry-After header's delay.

Limits and fair use

The API is free and has no key, so it has limits that keep it available for everyone:

Per PDF	20 MB, 50 pages, 120 seconds of conversion time
Per address	20 conversions per 10 minutes and 100 per day, shared with the website
At once	1 conversion in progress per address; send the next file when the previous response arrives

Send files one after another, not in parallel. Scanned PDFs (images with no text inside) are not supported.