PDF to Markdown & Text Converter for LLM & RAG
Pricing
from $1.50 / 1,000 pdf converteds
PDF to Markdown & Text Converter for LLM & RAG
Convert PDF URLs in bulk to clean Markdown or plain text for LLMs, RAG and search indexing: headings, paragraphs, lists, best-effort tables, plus title, author, date, page count and word count. No OCR, so scanned image-only PDFs are skipped and never charged.
Pricing
from $1.50 / 1,000 pdf converteds
Rating
0.0
(0)
Developer
LiveFree Analytics
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
18 days ago
Last modified
Categories
Features
Share
PDF to Text & Markdown: Bulk PDF Extractor for AI & RAG
Paste PDF links, or a page that links to PDFs. Get clean, searchable text back, as Markdown with headings or plain text, plus the title, author, dates, page count and word count. Ready for ChatGPT, Claude, a vector database or a spreadsheet.
Who it's for
- AI builders: load reports, manuals, papers and policies into RAG pipelines
- Researchers and analysts: pull text from dozens of reports or filings at once
- Compliance and procurement: search across a pile of public documents
- AI agents: let an assistant read any PDF link through the Apify MCP server
What you get (real output row, text shortened)
{"url": "https://www.irs.gov/pub/irs-pdf/fw9.pdf","status": "OK","title": "Form W-9 (Rev. March 2024)","subject": "Request for Taxpayer Identification Number and Certification","creationDate": "2024-03-06T13:18:13.000Z","language": "en-US","pageCount": 6,"wordCount": 6348,"markdown": "# Form W-9(Rev. March 2024)\n\n## Request for Taxpayer Identification Number and Certification\n\n...","charged": true,"durationMs": 927}
Pages are separated by ---. Very long documents are saved in full to the key-value store (fullTextKey).
Try it in 1 minute
-
Paste PDF links into Start URLs, or a web page and turn on Find PDF links.
-
Choose Markdown, Text or Both.
-
Click Start.
{ "startUrls": [{ "url": "https://arxiv.org/pdf/1706.03762" }], "outputFormat": "markdown" }
Popular setups
| Goal | Settings |
|---|---|
| Every PDF linked from a page | findPdfLinks: true, maxPdfLinksPerPage: 50 |
| Just the first pages | pageRange: "1-3" |
| Per-page text for citations | includePages: true |
| Strict budget | maxChargedDocuments: 100 |
| Big files | maxFileSizeMb: 100 (default) |
Speed
It reads the PDF's text layer directly, with no browser. In our tests a 15-page arXiv paper took under 1 second and a 142-page IRS publication about 3 seconds.
Price
$3.00 per 1,000 documents on Apify's Free plan ($2.50 Starter, $2.00 Scale, $1.50 Business). It's per document, not per page. Compute is included.
- No start fee from us. Only Apify's standard minimum start event (a fraction of a cent).
- Free: scanned PDFs with no text (
NO_TEXT_LAYER), password-protected (ENCRYPTED), broken (CORRUPT), not a PDF (NOT_PDF), too large (TOO_LARGE) and failed downloads. - If no document could be converted, the run fails with a clear message and costs nothing.
All settings
| Field | Default | What it does |
|---|---|---|
startUrls | – (required) | PDF links, or pages that link to PDFs |
outputFormat | markdown | markdown, text or both |
pageRange / maxPagesPerDoc | – / 0 (all) | Limit pages |
includePages | false | Also output text per page |
removeHeadersFooters | true | Drop repeated headers and footers |
findPdfLinks / maxPdfLinksPerPage | false / 50 | Discover PDFs on a web page |
maxChargedDocuments / maxFileSizeMb | 0 (no limit) / 100 | Budget and size limits |
includeFailedDocuments | true | Keep free failure rows |
maxConcurrency / requestTimeoutSecs / maxRetries | 4 / 120 / 1 | Speed and reliability |
userAgent / proxyConfiguration | – / off | Advanced |
Use it from code, no-code tools or AI agents
- API:
POST https://api.apify.com/v2/acts/blond_terrarium~pdf-to-text-markdown/run-sync-get-dataset-items?token=YOUR_TOKEN - LangChain / LlamaIndex dataset loaders, plus Make, Zapier and n8n
- AI agents (MCP):
https://mcp.apify.com?tools=blond_terrarium/pdf-to-text-markdown. Ask "Read this PDF and list its key dates".
FAQ
Scanned PDFs? Not supported (no OCR). They come back as NO_TEXT_LAYER and are free. Use an OCR actor for those.
Can I upload files from my computer? Use a public link. Any URL works, including Apify key-value store or cloud-storage links.
Tables? The text comes through in reading order. Complex tables may not keep their columns.
Support? Use the Issues tab. We read every one.
Related tools
Website to Markdown for LLMs · Google News Scraper
Built on pdf.js (Apache-2.0). Built and maintained by LiveFree Analytics.