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_COMPLETE
Generates a response (completion) from text or an image using a supported language model. You can provide: A text prompt, to generate a response from the model.
SnowConvert AI - Preview Features Settings
The Preview Features Settings in SnowConvert AI allow you to enable conversions that utilize Snowflake Public Preview features.
AI_COMPLETE structured outputs
AI_COMPLETE lets you supply a JSON schema or type literal that completion responses must follow, producing structured output. Structured output reduces the need for post - processing in your AI data pipelines and enables seamless integration…
Extract information with Document AI
Feature — Generally Available Available to accounts in AWS, Microsoft Azure, and Google Cloud commercial regions, with some exceptions.
Document AI PREDICT() function fails with "Invalid async token for External Function"
CAUSE: This error is caused by the total length of the name used for the DocumentAI build.
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_COMPLETE (Single image)
Generates a response (completion) for a text prompt using a supported language model. Syntax The function has two required arguments and four optional arguments. The function can be used with either positional or named argument syntax.
SnowConvert AI - Migration Assistant - Troubleshooting
SnowConvert AI Migration Assistant uses Snowflake Cortex AI to generate suggestions and explanations using Large Language Models (LLMs). These models can make mistakes, so please review each output thoroughly and carefully before applying…