Snowflake Cortex currently offers the following serverless functions:
Answer Extraction (in private preview): Extracts information from unstructured data.
Sentiment Detection (in private preview): Detects sentiment of text across a table.
Text Summarization (in private preview): Summarizes long documents for faster consumption.
Code Generation (in private preview): Generates code in a variety of programming languages.
Translation (in private preview): Translates text from one language to another.
Question Answering (in private preview): Answers questions about data.
Snowflake is planning to add more serverless functions to Cortex in the future, including functions for:
Image classification: Classifies images into different categories.
Object detection: Detects objects in images.
Natural language inference: Determines whether a hypothesis is entailed by a premise.
Paraphrasing: Generates paraphrases of text.
Creative text generation: Generates creative text formats of text content, like poems, code, scripts, musical pieces, email, letters, etc.
Snowflake Cortex serverless functions can be called from SQL or Python code. This makes it easy to integrate them into existing data pipelines and applications.
Here are some examples of how Snowflake Cortex serverless functions can be used:
A company can use the Answer Extraction function to extract customer information from support tickets.
A financial services company can use the Sentiment Detection function to detect sentiment in social media posts about its products and services.
A news organization can use the Text Summarization function to summarize long news articles.
A software company can use the Code Generation function to generate code stubs for new features.
A translation agency can use the Translation function to translate documents into different languages.
A customer service team can use the Question Answering function to answer customer questions about products and services.
Overall, Snowflake Cortex serverless functions offer a powerful and versatile way to use generative AI in the Data Cloud.