PDF to Text & Markdown — Extract Text, Pages, RAG Chunks
Pricing
$3.00 / 1,000 pdf extracteds
PDF to Text & Markdown — Extract Text, Pages, RAG Chunks
Extract text from PDF URLs as plain text, Markdown with headings, per-page rows or RAG-ready chunks, plus metadata. Fast, no start fee, scanned PDFs without a text layer are flagged and not charged.
Pricing
$3.00 / 1,000 pdf extracteds
Rating
0.0
(0)
Developer
Bruno Scherer Oliveira
Maintained by CommunityActor stats
0
Bookmarked
2
Total users
1
Monthly active users
3 days ago
Last modified
Categories
Security
Limited permissionsShare
Turn PDF links into clean text in seconds. You get plain text, Markdown with headings, one row per page, or overlapping chunks ready for embeddings and RAG, plus document metadata (title, author, dates, page count).
There's no start fee and one simple price per PDF. Scanned PDFs without a text layer are flagged and not charged, and neither are failed downloads.
Use cases
- Feed reports, papers, contracts and manuals into LLMs, vector databases and RAG pipelines
- Bulk-convert PDF libraries to searchable text or Markdown
- Pull text from specific pages, e.g. the first 3 pages of every annual report
- Index PDFs linked from your own sites or datasets
Input
| Field | What it does |
|---|---|
| PDF URLs | Direct links to PDF files |
| Output layout | document (one row per PDF), pages (one row per page) or chunks (RAG chunks) |
| Markdown | Adds Markdown with headings detected from font sizes |
| Page range | e.g. 1-5,8,10- |
| Max pages per PDF | Cap long documents |
| Chunk size / overlap | For the chunks layout. Chunks break at paragraph, line or sentence boundaries |
| Max file size | Larger files are skipped and not charged (default 50 MB) |
| Parallel downloads | 1–10 at once. Requests to the same host are spaced 1 second apart |
Example: RAG chunks from a paper.
Output
In the pages layout each row adds page, and in the chunks layout each row adds chunkIndex and chunkCount.
Statuses:
ok: text extracted, charged once per PDFno_text_layer: a scanned or image-only PDF, not charged (it needs OCR, which this Actor doesn't do)failed: download error, not a PDF, or too large, with the reason inerror, not charged
Good to know
- Extraction uses Mozilla's pdf.js, the engine inside Firefox. Text comes out in reading order per line. Complex multi-column layouts and tables come out as text lines, not as a layout.
- Markdown headings are inferred from font size, so they're a helpful structure rather than a perfect outline.
- Only process PDFs you're allowed to access and use.
Pricing
One charge per successfully extracted PDF, whatever the layout or page count. No start fee.
