Snowflake AI SQL User Guide: Extract text from documents into accurate, queryable formats using Snowflake’s AI_PARSE_DOCUMENT.

Snowflake Community is a platform for users of the Snowflake cloud data platform to share knowledge, ask questions, and collaborate. Readers can learn about cloud data warehousing, data analytics, and data engineering best practices. With forums, user groups, and community events, Snowflake Community provides resources for Snowflake users to connect and learn from each other.

Snowflake Community

AI_PARSE_DOCUMENT is a Snowflake Cortex AI SQL function for extracting text, structure, and images from documents (PDF, DOCX, PPTX, and common image formats). It supports four main modes: OCR for plain text extraction, LAYOUT for structured Markdown output preserving tables, page filtering to target specific page ranges, and image extraction returning base64-encoded embedded images. The guide walks through setup (database, schema, stage, warehouse), SQL syntax for each mode, and how to parse the returned JSON using Snowflake's colon notation and LATERAL FLATTEN. It also shows how to combine AI_PARSE_DOCUMENT with AI_EXTRACT to analyze extracted images and with AI_COMPLETE for document summarization.

Snowflake AI SQL User Guide: AI_PARSE_DOCUMENT

2. Layout Mode (Structured Markdown with Tables)

Get Douglas Day’s stories in your inbox

3. Page Filtering (Process Specific Pages Only)

4. Image Extraction (Extract Embedded Images)