PDFs in. AI-ready text out.
Turn PDF links into clean full text, per-page text, metadata and token-sized chunks you can embed straight into a vector database.
Built for the AI pipeline.
Clean full text
Readable text with layout noise removed, plus word and page counts.
Per-page output
Keep page numbers for citations and source links in your answers.
Ready-made chunks
Overlapping chunks at the size you choose, ready to embed.
Metadata
Title, author, dates and page count from every document.
Fast and parallel
Hundreds of PDFs per run, with failures reported per file.
Fits your stack
API, LangChain, LlamaIndex, Make, Zapier and n8n friendly JSON.
Three steps. That's it.
Add PDF links
Paste URLs or point it at files in storage.
Choose chunk size
Pick a chunk size and overlap, or keep the sensible defaults.
Embed
Pull the JSON into your vector store or download it as a file.
A fifth of a cent per document.
No start fee and nothing to maintain. 1,000 PDFs cost about $2, and failed downloads are free.
- Full text, per-page text and chunks
- Document metadata
- Configurable chunk size and overlap
- API and integrations
Do I need to code?
No. Open the tool on Apify, paste your input, press Start and download the results as CSV, Excel or JSON. Developers can also call it through the Apify API, Make, Zapier or n8n.
How do I pay?
Through your Apify account. Apify gives every free account monthly platform credit, which is usually enough to try the tool at no cost.
Does it read scanned PDFs?
It extracts the text layer that most digital PDFs have. Scanned images without a text layer return an error and are not charged.
You might also like
Website Contact Finder
A list of websites in, one clean row per company out: emails, phones and every social profile.
Domain Email & DNS Checker
Check thousands of domains: can they receive email, who hosts their mail, SPF and DMARC status.
Bank Statement & Invoice PDF to CSV
Bank statement PDFs become clean transaction rows. Invoices become fields and line items.