Semantic Studio reaches GA for AI-powered semantic view authoring

Use cases

Imagine you need to structure the business meaning of your data so AI tools and users can query it easily. With Semantic Studio, you create and manage semantic views in a unified authoring environment inside Snowsight.

• Chat with the CoCo agent to create and refine semantic views using natural language.
• Edit YAML files directly and integrate them with Git version control.
• Deploy local definitions directly to active Snowflake objects.

How to use

Before you begin, verify that a supported large language model is available in your account by running this command in a SQL Worksheet:

SHOW MODELS LIKE 'claude-4-sonnet' IN SNOWFLAKE.MODELS;

What happens in this code:

1. The command queries the available models within the Snowflake scope.
2. It returns the list matching the specified pattern, ensuring your account's AI assistant is ready for use.

Limitations and requirements

• Generally available (GA).
• Requires a role with privileges such as CREATE SEMANTIC VIEW on the target schema, USAGE on the database and schema, and SELECT on the underlying tables.
• Requires at least one supported large language model enabled in the account.