Urdu OCR API
Read Urdu text, including Nastaliq print, where letters slope and stack along each word.
Result
Your Markdown or JSON shows up here.
What makes Urdu hard for OCR
- Nastaliq letters stacked diagonally
- Most engines are trained on Naskh, not Nastaliq
- Right-to-left text with left-to-right numbers
Script: Perso-Arabic script, usually Nastaliq.
Call it from your code
curl https://api.ocrrouter.ai/v1/ocr \
-H "Authorization: Bearer $OCRROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google/gemini-flash",
"file_url": "https://example.com/invoice.pdf",
"format": "markdown"
}'Best models for Urdu text
1Gemini FlashGeneral vision model with strong handwriting and layout understanding at a low price.
google/gemini-flash3 credits / page2GPTGeneral vision model. Good at messy scans and following extraction instructions.openai/gpt11 credits / page3Google Document AIGoogle Cloud OCR with layout parsing and specialised processors.google/document-ai3 credits / page