Cookbooks
Patterns for the OCRQueen API
End-to-end recipes for the most common workflows. Copy, paste, adapt. New to the API? Start with the Quickstart.
Ingest documents into a RAG pipeline
Chunk an OCRQueen extraction into LLM-friendly pieces with provenance + page metadata preserved.
Extract patents (figures + claims + references)
Use profile=advanced to capture diagram graphs, image alt-text, and cross-link reference numerals.
Batch processing with webhooks
Submit thousands of documents in parallel without polling — receive results on your webhook receiver.