Legal OCR
Make scanned pleadings, court orders, deeds and discovery productions searchable, with case numbers, parties and dates pulled out for your matter list.
PDF, PNG, JPG, WebP, HEIC. Up to 10 files at onceNo file?
What makes legal documents hard to read
- Line-numbered pleading paper and margin text
- Bates stamps, exhibit stickers and redaction boxes
- Thousands of pages per production
Fields coming with document_type
{"document_kind": "court order","court": "Superior Court of California, County of Alameda","case_number": "RG25-118842","parties[]": "role, name","filed_date": "2026-08-19","judge": "Hon. M. Delgado","bates_range": "DEF-004120 to DEF-004187","text": "full text as Markdown"}Best models for legal documents
1ClaudeGeneral vision model with careful reading of long, dense documents.25 credits / page
anthropic/claude25 credits / page2Mistral OCRFast document OCR that returns clean Markdown with tables and embedded images.25 credits / pagemistral/mistral-ocr25 credits / page3Reducto ParseHigh-accuracy parsing for complex financial and legal documents, with citations.20 credits / pagereducto/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": "anthropic/claude",
"file_url": "https://example.com/invoice.pdf",
"format": "json"
}'