Creating semantic views¶

You can create a semantic view in Snowsight with Semantic View Autopilot, by writing code (SQL or YAML), or by importing a model you already maintain. Use this page to choose an approach; each option links to detailed instructions.

Create with Semantic View Autopilot¶

Semantic View Autopilot is the AI-assisted creation wizard in Snowsight. You reach it from Semantic Studio: in Workspaces, select Add new » Semantic View. The wizard offers several options to get started, and Snowflake recommends CoCo, which guides you through creating the view conversationally. See Semantic View Autopilot.

You can also seed the view from a file or from example queries:

• Import a Tableau workbook or data source. See Tableau ingestion feature support.
• Import a Power BI semantic model. See Power BI ingestion feature support.
• Upload a YAML specification, or provide example SQL queries. See Options for providing context.

After the view exists, refine and deploy it in Semantic Studio.

Create with code¶

Define semantic views as code for version control and automated deployment:

• Use DDL (SQL). See Using SQL commands to create and manage semantic views.
• Use YAML. See YAML specification for semantic views.
• Decide which to use. See YAML vs DDL authoring for semantic views.

Migrate an existing model¶

Move from the legacy stage-based semantic models to native semantic views. See Migrating from the legacy stage API to semantic views.