PDF Text Extractor

Extract clean text, metadata and page counts from any PDF URL — built for RAG pipelines, LLM ingestion and large document datasets. Give it a list of PDF links and it downloads each file, verifies it is a real PDF, and returns the full text, optional per-page text, page count, word count and document metadata (title, author, subject, creator, producer, creation date).

No API key. No login. Datacenter-proxy friendly (it operates on the PDF URLs you provide), so it runs cheaply on the smallest Apify plan.

Why use this actor

• RAG / LLM ingestion — turn PDFs into clean text ready to chunk and embed.
• Document data extraction — pull text + metadata from invoices, papers, manuals, reports, contracts and public filings at scale.
• Research — batch-download and read arXiv, bioRxiv, SSRN or .gov PDFs.
• Archiving & indexing — build a searchable text index of a PDF collection.

Features

• Accepts any list of direct PDF URLs (redirects followed automatically).
• Full concatenated text (capped at 200,000 characters) per document.
• Optional per-page text array (includePages) for page-accurate chunking.
• Document metadata: title, author, subject, creator, producer, creation date.
• Page count and word count for every PDF.
• Robust: non-PDF URLs are logged and skipped; malformed PDFs never crash a run.
• One dataset item per PDF, ready to export as JSON, CSV, Excel or feed an API.

Input

Field	Type	Description
mode	select	Operation mode. Currently extract.
urls	array	Direct links to PDF files to extract.
includePages	boolean	Add a per-page text array (default false).
maxItems	integer	Max PDFs to process, 1–1000 (default 100).

Example input

{
  "mode": "extract",
  "urls": [
    "https://arxiv.org/pdf/1706.03762",
    "https://www.w3.org/WAI/ER/tests/xhtml/testfiles/resources/pdf/dummy.pdf"
  ],
  "includePages": false,
  "maxItems": 100
}

Output

Each PDF becomes one dataset record:

{
  "url": "https://arxiv.org/pdf/1706.03762",
  "final_url": "https://arxiv.org/pdf/1706.03762",
  "title": null,
  "author": null,
  "subject": null,
  "creator": "LaTeX with hyperref",
  "producer": "pdfTeX-1.40.25",
  "creation_date": "2024-04-10T21:11:43+00:00",
  "page_count": 15,
  "text": "Attention Is All You Need ...",
  "pages": null,
  "word_count": 6123,
  "source": "pdf-extractor",
  "scraped_at": "2026-07-21T16:20:00Z"
}

Pricing

Cheap by design: about $1 per 1,000 PDFs plus Apify platform usage. Text extraction is pure-Python (no browser, no paid proxy), so runs are fast and the compute cost stays low.

Related actors

• Smart Article Extractor
• Website Content Crawler
• HTML to Markdown Converter
• Wayback Machine Scraper

Keywords

pdf text extractor, extract text from pdf, pdf to text, pdf scraper, pdf parser, pdf metadata extractor, pdf page count, rag pdf ingestion, llm document ingestion, document dataset, pdf data extraction, arxiv pdf, ocr alternative, batch pdf extraction, pdf api.