PDF Text Extractor — Text, Markdown, Tables and OCR

Extract text from PDF files: full text, page-by-page text, clean Markdown, tables and document metadata. Scanned pages without a text layer are read with OCR in 19 languages. Password-protected PDFs, page ranges and RAG chunks are supported. $0.003 per PDF + $0.0001 per page, no fee per run, no login, no API key.

It is the Document Text Extractor engine with the form set up for PDFs; anything that works there works here, including Word, Excel, PowerPoint, EPUB, HTML, CSV and TXT links.

What it is for

What you give it	What you get back
A PDF link, an uploaded PDF or base64 data	Full text, per-page text, Markdown, metadata
A report or paper with tables	Each table as Markdown and as a row array
A scanned PDF	OCR text for the pages that have no text layer
A password-protected PDF	The same output, with the password you pass
Hundreds of PDF links	One row per PDF, same fields for all of them

An empty call ({}) reads a two-page sample PDF, so you can see the output shape before sending your own files.

What data does it return?

Field	Type	Example
source / fileName	string	https://arxiv.org/pdf/1706.03762
pageCount / pagesProcessed	integer	15 / 15
title, author, subject, creator, producer	string	Annual Operations Report 2026
documentCreatedAt / documentModifiedAt	string	2026-09-05T00:07:28+08:00
text	string	the whole document as plain text
markdown	string	headings, lists and tables kept
pages	array	[{"number": 1, "text": "…", "markdown": "…", "ocr": false}]
tables / tableCount	array / integer	`[{"page": 3, "markdown": "
chunks / chunkCount	array / integer	when Split into RAG chunks is on
wordCount, charCount, ocrPageCount	integer	6002, 39811, 0
status / errorCode / errorMessage	string	ok, or why a file failed
warnings	array	["2 scanned page(s) were not processed because the OCR page limit was reached."]

How much does it cost?

Event	Price	When it is charged
PDF processed	$0.003	Once per PDF that returned text
Page parsed	$0.0001	Per page of text
Scanned page read with OCR	$0.008	Only for pages with no text layer, and only when OCR found text

Job	Cost
1 PDF of 10 pages	$0.004
1,000 PDFs of 10 pages	$4.00
50 scanned PDFs of 5 pages, OCR	$2.18

There is no fee per run. A dead link, a corrupt file or a PDF with no readable text comes back as an error row and is never charged. Pages that already contain text are never OCR-ed. Switch OCR pages without a text layer off and scanned pages are skipped at no cost.

How to extract text from a PDF

1. Click Try for free and sign in to Apify.
2. Paste PDF links into PDF URLs, or use Upload a file.
3. Leave Markdown output on if the text goes to an LLM; switch on Split into RAG chunks for a vector database.
4. For scans in another language, pick the OCR language.
5. Press Start, then read the Output tab or export JSON, CSV or Excel.

⬇️ Input

{
  "urls": ["https://raw.githubusercontent.com/lergassy/apify-actor-assets/main/document-text-extractor/sample-report.pdf"],
  "includePages": true,
  "extractTables": true,
  "ocrScannedPages": true,
  "ocrLanguage": "eng"
}

Field	What it does
urls	Direct links to the PDF files (not to a viewer page)
documentFile	Upload one file from your computer
base64Files	Base64 strings or {"fileName", "data"} objects, for API and agent calls
includeMarkdown / includeFullText / includePages	Which text fields to return (all on by default)
extractTables	Detect tables and return them as Markdown and rows
chunkForRag, chunkSize, chunkOverlap	Split into retrieval chunks that keep their heading path
pageRange	Read only some pages, e.g. 1-5, 8 or 12-
password	Password for a protected PDF
ocrScannedPages, ocrLanguage, ocrMaxPagesPerDocument	OCR for pages without a text layer (on by default, capped at 20 pages per PDF)
maxConcurrency, maxFileSizeMb, timeoutPerDocumentSecs	Advanced limits

If a host refuses the download (rate limit, block), the Actor retries with a pause and then through a residential proxy before giving up on the file. You are not charged for that.

⬆️ Output

{
  "source": "https://…/sample-report.pdf",
  "format": "pdf",
  "pageCount": 2,
  "title": "Annual Operations Report 2026",
  "author": "Malina Data",
  "wordCount": 198,
  "ocrPageCount": 0,
  "markdown": "# Annual Operations Report 2026\n\n## Executive summary\n…",
  "pages": [{ "number": 1, "text": "Annual Operations Report 2026…", "ocr": false }],
  "tableCount": 1,
  "status": "ok"
}

Error rows

errorCode	Meaning
http-error / network-error / timeout	The file could not be downloaded, even after retries
too-large	Above the size limit (50 MB by default, up to 200 MB)
password-required	The PDF needs a password
corrupt-file	The file is damaged or not a PDF
no-text-content	No text found — a scan with OCR switched off, or an empty document
unsupported-format	Not a supported file type

Limitations, stated plainly

• OCR is built for printed text; handwriting and heavy stamps read poorly.
• Tables are detected from the PDF's layout. Tables drawn without lines, or split across pages, may come back merged or partial.
• At most 20 scanned pages per PDF are OCR-ed by default (adjustable up to 500), so a long scan cannot surprise you.
• Google Docs or Dropbox viewer links are web pages, not files — use the direct download link.

Use cases

• RAG and knowledge bases — Markdown and chunks with their heading path go straight into embeddings.
• Contract, invoice and report review — full text plus metadata and tables for search or an LLM.
• Research papers — per-page text and tables from arXiv and journal PDFs.
• Scanned archives — OCR only where needed, billed per scanned page.

🤖 For AI agents and LLM apps

{ "urls": ["https://example.com/report.pdf"] }

• One row per PDF; status is ok or error.
• markdown is the field to give a model; chunks is the field to embed.
• Files can be passed as base64; an empty call returns a sample result instead of an error.

Use via API

curl -X POST "https://api.apify.com/v2/acts/lergassy~pdf-text-extractor/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://example.com/report.pdf"]}'

from apify_client import ApifyClient
items = ApifyClient("YOUR_TOKEN").actor("lergassy/pdf-text-extractor").call(
    run_input={"urls": ["https://example.com/report.pdf"]}
).iterate_items()

Works the same from JavaScript, n8n, Make, Zapier and the Apify MCP server.

❓ FAQ

Does one row mean one PDF or one page?

One PDF. Pages are inside the row in pages.

Can it read password-protected PDFs?

Yes, pass the password in PDF password.

What is the difference from Document Text Extractor?

None in the engine. Document Text Extractor is the main Actor, framed around every document format; this one is set up for PDFs, with per-page output and OCR on by default.

I only have images, not PDFs.

Use Image to Text — the same OCR engine, with confidence scores and word positions.

You might also like

Actor	What it does
Document Text Extractor	The main Actor: PDF, Word, Excel, PowerPoint, EPUB to Markdown and RAG chunks
OCR Text Extractor	Scanned PDFs and images with confidence scores and word boxes
Image to Text	Screenshots and photos to text

Also known as

PDF to text, PDF text extraction API, PDF parser, PDF to Markdown, extract text from PDF, PDF OCR, PDF scraper.