Source
Document AI
Document AI is a Snowflake AI feature that uses Arctic - TILT, a proprietary large language model (LLM), to extract data from documents.
AI_AGG
Reduces a column of text data using a natural language instruction. For example, AI_AGG(reviews, 'Describe the most common complaints mentioned in the book reviews') will return a summary of user feedback.
AI_SUMMARIZE_AGG
Summarizes a column of text data. For example, AI_SUMMARIZE_AGG(churn_reason) will return a summary of the churn_reason column. Unlike AI_COMPLETE and SUMMARIZE (SNOWFLAKE.CORTEX), this function supports datasets larger than the maximum…
AI_CLASSIFY
AI_CLASSIFY is the updated version of CLASSIFY_TEXT (SNOWFLAKE.CORTEX). For the latest functionality, use AI_CLASSIFY. Classifies text or images into categories that you specify.
AI_COMPLETE (Prompt object)
Generates a response (completion) for a prompt object. The prompt object references one or more columns containing text or image data. Syntax The function can be used with either positional or named argument syntax.
AI_COUNT_TOKENS
AI_COUNT_TOKENS is the updated version of COUNT_TOKENS (SNOWFLAKE.CORTEX). For the latest functionality, use AI_COUNT_TOKENS. Returns the number of tokens in a prompt for the specified large language model or task - specific function.
AISQL AI_PARSE_DOCUMENT
AI_PARSE_DOCUMENT is a Cortex AI SQL function that extracts text, data, and layout elements from documents with high fidelity.
Extract information with Document AI
Feature — Generally Available Available to accounts in AWS, Microsoft Azure, and Google Cloud commercial regions, with some exceptions.
AI_FILTER
Classifies free - form prompt inputs into a boolean. Currently supports both text and image filtering.
Known limitations to Document AI
The Document AI model returns answers that are up to 512 tokens long (about 320 words) per question. The model for table extraction returns answers that are up to 2048 tokens long. For internal stages, Document AI supports using…