Insurance claim OCR API
Extract policy, claimant, incident and amount details from insurance claim forms and supporting documents.
Result
Your Markdown or JSON shows up here.
What makes insurance claims hard to read
- Long forms with handwritten narratives
- Mixed bundles of forms, invoices and photos
- Amounts spread across several documents
What you can pull out
insurance-claim fields
{
"claim_number": "CLM-2026-004418",
"policy_number": "HOM-7781223",
"claimant": "Grace Thompson",
"incident_date": "2026-09-22",
"incident_type": "water damage",
"description": "Burst pipe in kitchen ceiling",
"claimed_amount": "8,450.00",
"attachments[]": "type, page"
}Best models for insurance claims
1Azure Document IntelligenceAzure layout model with tables, selection marks and prebuilt invoice and ID readers.
azure/document-intelligence15 credits / page2Gemini FlashGeneral vision model with strong handwriting and layout understanding at a low price.google/gemini-flash3 credits / page3Reducto ParseHigh-accuracy parsing for complex financial and legal documents, with citations.reducto/parse20 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": "azure/document-intelligence",
"file_url": "https://example.com/invoice.pdf",
"format": "json"
}'