Document Extraction API
Integrate AI document extraction into your product or workflow with a simple REST API. Extract structured JSON from invoices, contracts, purchase orders, receipts, and scanned PDFs — no ML infrastructure required.
What Paperloom extracts
POST a document PDF or image and receive structured JSON back. Simple authentication, predictable response format, and webhook support for async workflows.
One API endpoint handles invoices, contracts, purchase orders, receipts, bank statements, and general PDFs — with auto-detection of document type.
Every response returns field names, extracted values, confidence scores, and source bounding boxes in a consistent JSON schema.
Every extracted field includes a confidence score so your application can decide whether to auto-approve or route to human review.
Register a webhook endpoint to receive extraction results asynchronously for batch processing and document pipeline workflows.
The API handles documents in Arabic, French, Spanish, German, Chinese, Japanese, and other languages without any configuration parameters.
How it works
Authenticate with your API key from the Paperloom dashboard.
POST your document (PDF, JPEG, PNG, or TIFF) to the /api/extract endpoint.
Specify the document type or let Paperloom auto-detect invoice, contract, PO, receipt, or general PDF.
Receive a structured JSON response with field values, confidence scores, and source locations.
Route low-confidence fields to your own review UI or use Paperloom's built-in review queue.
Use cases
Finance software companies embed Paperloom's API to add invoice extraction capabilities to their products without building ML pipelines.
Developers use the document extraction API to build document-to-ERP workflows that auto-populate fields from uploaded invoices and POs.
DMS platforms integrate the API to automatically extract metadata from uploaded documents and index them for search and compliance.
Operations teams build custom document processing pipelines that extract, validate, and route data from uploaded documents without manual intervention.
No setup. No credit card. 20 free credits on sign-up.
Get Started FreeA document extraction API accepts a PDF or document image as input and returns structured data — field names, values, and confidence scores — as JSON. Developers use it to add document extraction capabilities to their products without building or maintaining ML models.
The API supports invoices, contracts, purchase orders, receipts, bank statements, bills of lading, and general PDFs. Pass the document type as a parameter or let the API auto-detect it.
The API returns structured JSON with field names, extracted string or numeric values, confidence scores (0–1), and optional source coordinates (bounding boxes) on the original document.
Rate limits depend on your plan. Free accounts can extract up to 20 documents. Paid plans include higher document limits with options for bulk processing and dedicated infrastructure for high-volume use.
Yes. For large PDFs or batch processing, the API supports async mode where you submit a document, receive a job ID, and either poll for results or register a webhook for callbacks.
AI Invoice Extraction
Automate invoice data extraction for AP teams.
Contract Data Extraction
Extract clauses and obligations from contracts.
Purchase Order Extraction
Automate PO data capture for procurement teams.
Receipt Extraction
Extract merchant, tax, and line items from receipts.
Bank Statement Extraction
Parse transactions from bank statement PDFs.
Azure Document Intelligence Alternative
Compare Paperloom vs Azure Document Intelligence.