Document Parsing API
Extract clean text from PDFs, scanned documents, images and DOCX files with one API call. Ready for search, storage or your LLM pipeline.
curl -X POST https://api.parsinga.com/auth/fileconversion \
-H "Authorization: Bearer $PARSINGA_API_KEY" \
-F "image=@document.pdf"{
"text": "INVOICE #INV-2041\nDate: 14 March 2026\nBill to: Acme Supplies\n..."
}Half a cent a page at volume.
Pay per page with no monthly plan or minimum. After your first 1,000 pages, every page costs $0.005.
- 100 free parses to start
- Only successful parses are charged
- No monthly plan or minimum
| 1,000 pages | $10 |
| 10,000 pages | $55 |
| 100,000 pages | $505 |
| 1,000,000 pages | $5,005 |
See a document parsed to text.
What teams use it for.
LLM and RAG pipelines
Clean text from any file, ready to chunk, embed and send to a language model.
Search and indexing
Make scanned archives and uploaded files searchable by their full content.
Invoices, contracts and forms
Get the text first, then extract the fields you need with a custom parser.
Digitising paper records
Turn scans and phone photos of documents into editable, storable text.
Document parsing questions.
Parsing resumes? See the Resume Parser API.
What does the document parsing API return?
The full text of the document as JSON. Send a PDF, scanned document, image or DOCX file and get its text back, ready to store, search or pass to an LLM.
How much does document parsing cost?
$0.01 per page for 1 – 1,000 pages, $0.005 per page for 1,001 and above pages. New accounts get 100 free parses, with no monthly plan or minimum.
Which file types are supported?
PDF (digital and scanned), DOC, DOCX and common image formats such as phone photos of documents.
Can it extract specific fields such as invoice totals?
This endpoint returns text. For structured fields from invoices, contracts, claims or any other document type, we build a custom parser on your samples and validate it by hand before it goes live.
Do you store documents or use them for training?
No. Documents are not stored once parsing is complete, and your data is never used to train LLM models.
Parse your first documents free.
100 free parses to start, then from $0.005 a page.