May 04, 2026: AI_PARSE_DOCUMENT OCR quality improvements

The Snowflake Cortex AI_PARSE_DOCUMENT function has launched significant quality improvements for OCR mode, building on the strong OCR performance already available in AI_PARSE_DOCUMENT and delivering even higher accuracy and faster processing for enterprise-scale document workflows that require efficient, high-quality text extraction.

Key improvements include:

  • Higher OCR accuracy: Up to a 25% improvement in character and word recognition, significantly reducing errors in low-quality scans and complex document types, leading to more reliable downstream parsing.

  • Enhanced multilingual support: Up to 20% improvement in recognition accuracy across supported languages, enabling more consistent performance in global document processing use cases.

  • Improved handling of multi-column documents: Better extraction accuracy for multi-column and other text-dense document formats, improving reading order and downstream usability.

For more information, see Parsing documents with AI_PARSE_DOCUMENT.