Mortgage document OCR
Classify and read loan packages, including the Loan Estimate, Closing Disclosure, note and deed of trust, and pull out the loan terms underwriters check.
PDF, PNG, JPG, WebP, HEIC. Up to 10 files at onceNo file?
What makes mortgage documents hard to read
- Packages of 100+ pages with many document types
- Fixed CFPB forms next to free-form lender pages
- Signatures, initials and notary stamps on every page
Fields coming with document_type
{"document_kind": "Closing Disclosure","borrowers[]": "Emily Carter, James Carter","property_address": "2208 Willow Bend Dr, Raleigh NC 27615","loan_amount": "384,000.00","interest_rate": "6.375%","loan_term": "30 years","monthly_principal_interest": "2,395.63","closing_date": "2026-10-30"}Best models for mortgage documents
1Reducto ParseHigh-accuracy parsing for complex financial and legal documents, with citations.20 credits / page
reducto/parse20 credits / page2ClaudeGeneral vision model with careful reading of long, dense documents.25 credits / pageanthropic/claude25 credits / page3Azure Document IntelligenceAzure layout model with tables, selection marks and prebuilt invoice and ID readers.60 credits / pageazure/document-intelligence60 credits / pageCall it from your code
curl https://api.ocrrouter.ai/v1/ocr \
-H "Authorization: Bearer $OCRROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "reducto/parse",
"file_url": "https://example.com/invoice.pdf",
"format": "json"
}'