← Back to the tool

PDF API

No API key. No account. No sign-up. Merge, compress, split, rotate, convert and OCR PDFs over plain HTTP. Post a file, get an id, download the result.

Last updated: 1 September 2026

This is the same engine the ScanReviver website runs on. It is free to call within the limits below, and there is nothing to register for.

How a call works

Every job is two steps. You POST your file and get a small JSON object back with an id. You then GET the result using that id. The reason for the split is that one job can produce more than one output: a merged PDF, a single page, or a ZIP of images.

# step 1 — send two PDFs to be merged
$ curl -s -F "[email protected]" -F "[email protected]" \
    https://scanreviver.com/api/pdf-merge
{"id":"5ea56646f09146979eabc2faecd0f889","bytes":72597,"files":2}

# step 2 — download the merged file
$ curl -s -o merged.pdf \
    https://scanreviver.com/api/pdf-file/5ea56646f09146979eabc2faecd0f889

Endpoints

All of these take multipart/form-data and return JSON containing an id.

Endpoint	Send	Then fetch from
POST /api/pdf-merge	files, one field repeated per PDF	/api/pdf-file/{id}
POST /api/pdf-compress	file, plus level (good by default)	/api/pdf-file/{id}
POST /api/pdf-split	file	/api/pdf-split-file/{id}
POST /api/pdf-to-images	file	/api/pdf-zip/{id}, or one page at a time from /api/pdf-page/{id}/{n}
POST /api/pdf-to-word	file	/api/pdf-word/{id}
POST /api/pdf-to-excel	file	/api/pdf-excel/{id}
POST /api/pdf-to-ppt	file	/api/pdf-ppt/{id}
POST /api/ocr-text	file	nothing — the text comes back in the same response

The site has more endpoints than these, covering rotating, cropping, page numbers, watermarks, forms and signing. They follow the same two-step shape. If you need one that is not listed here, ask on the contact page and we will document it properly rather than let you guess.

OCR returns text directly

/api/ocr-text is the one that does not need a second call, because the result is text rather than a file:

$ curl -s -F "[email protected]" https://scanreviver.com/api/ocr-text
{"text":"Invoice 2026 total 42 euro","pages":1,"total_pages":1,"chars":26}

Limits

Maximum per file	250 MB
Maximum per job	1000 MB
Pages per job	30
Jobs at once	1 per IP address
Jobs per day	No cap
Result stays available	45 minutes

The 45 minutes matter: download the result in the same run of your script. An id you saved yesterday will not work, and that is deliberate — nothing of yours sits on our disk longer than it has to.

Errors

Errors come back as JSON, not as a file, so check the status code before writing the body anywhere.

Status	Meaning
400	The file cannot be used for this operation, or the job id is unknown or older than 45 minutes. The message says which.
413	Too big, or too many pages.
422	A required form field is missing.
429	You already have a job running from this address.

A useful one to expect: sending a scanned PDF to the Word converter gives a 400 telling you to run OCR first, because there is no real text in the file to convert.

Python, end to end

import requests

BASE = "https://scanreviver.com"

r = requests.post(f"{BASE}/api/pdf-merge",
                  files=[("files", open("a.pdf", "rb")),
                         ("files", open("b.pdf", "rb"))],
                  timeout=180)
r.raise_for_status()
job = r.json()["id"]

out = requests.get(f"{BASE}/api/pdf-file/{job}", timeout=180)
out.raise_for_status()
open("merged.pdf", "wb").write(out.content)

Using it in your own project

Build on it, including in something you charge for. Two things we ask, neither enforced: keep to one job at a time so everyone gets a turn, and if your project lists what it uses, link back to scanreviver.com. That is what keeps this free.

Planning real volume? Say hello on the contact page first. We would rather hear from you than throttle you.

Questions

Terms are on the terms page, and what happens to your data is on the privacy page.