PDF to Text & AI Chunks Extractor
Pricing
$2.00 / 1,000 pdf extracteds
Go to Apify Store
PDF to Text & AI Chunks Extractor
Turn PDF links into clean text, per-page text, metadata and ready-to-embed chunks for RAG and AI agents. No start fee: pay a small price per PDF.
Pricing
$2.00 / 1,000 pdf extracteds
Rating
0.0
(0)
Developer
Saksham Singh
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 hours ago
Last modified
Categories
Features
Limited permissions
Share
Give it PDF links and get back, for each file:
- 📝 Clean full text (hyphenation fixed, extra spaces removed)
- 📄 Per-page text, so you can cite page numbers
- 🧩 AI-ready chunks with overlap, ready to embed for RAG, vector databases and AI agents
- 🏷️ Metadata: title, author, subject, creator, producer, creation date, page count, word count
- 🔎 A
likelyScannedflag for image-only PDFs that need OCR
Why this one
- No start fee. You pay a small flat price per PDF extracted. Failed downloads are free.
- Built for AI pipelines. Choose chunk size and overlap in words; each chunk has an index and start position.
- Fast. Pure text extraction, no browser, so runs finish in seconds and cost little.
Input
| Field | Description |
|---|---|
pdfUrls | Direct links to PDF files |
includePages | Add per-page text (default on) |
chunkSize / chunkOverlap | Chunk length and overlap in words (0 turns chunking off) |
maxPages | Only read the first N pages (0 = all) |
maxFileSizeMb | Skip files bigger than this |
Output example
{"url": "https://arxiv.org/pdf/1706.03762","title": "Attention Is All You Need","pageCount": 15,"wordCount": 6094,"text": "Attention Is All You Need ...","pages": [{ "page": 1, "text": "..." }],"chunks": [{ "index": 0, "startWord": 0, "text": "..." }]}
Export to JSON, CSV or Excel, or call it from the API, Make, Zapier, n8n, LangChain or an MCP-enabled AI agent.
Limits
Scanned PDFs without a text layer return little text (flagged with likelyScanned). Password-protected PDFs are not supported. Only process documents you have the right to use.