M↓ View as Markdown ↗

Creating Models

You create Models in Studio from the Create Model dialog. It offers three methods, so you can let Steward do the work, start from a warehouse table, or write your own SQL.

Availability

• Who can do this: members with the Edit capability in the active Workspace.

Open the Create Model dialog

1. In Studio, click Models in the sidebar.
2. Click + (New model) at the top of the Models list. The Create Model dialog opens.
3. Choose a method.

The dialog's Learn more about creating models link opens this page.

Choose a method

Method	What it does	Use it when
With Steward (Recommended)	Steward analyzes your schema, researches business context, and drafts a complete Model.	You want help deciding what the Model should contain, or you want relationships to other Models proposed as well.
From Table	Start from auto-detected dimensions and measures, then refine yourself.	The Model maps cleanly to one warehouse table and you want a quick starting point.
From SQL	Define a custom SQL query for complex transformations or multi-table joins.	The data you want to model doesn't exist as a single table.

• Creating Models with Steward
• Creating Models from Table
• Creating Models from SQL

What to know before you start

• You need a data connection. Every Model is built on one data connection in your Workspace. See Connect a data source.
• Names are lowercase snake_case. Model names start with a letter and use only lowercase letters, digits and underscores (for example orders, line_items). They must be unique among the Models in your Workspace.
• The data source is fixed. You can't change a Model's table or SQL query after you create it. To use a different source, create a new Model.
• AI drafts the first version. With From Table and From SQL, clicking Create model sends the column names and data types, up to 10 sample rows and (for From SQL) your query to the language model your organization uses. AI drafts the description, guidance, grain, dimensions and measures, saves the Model and opens it for you to review. With With Steward, Steward proposes the Model and, by default, nothing is saved until you accept the proposed edits.
• Relationships come later. From Table and From SQL don't create relationships. Add them on the Model's Relationships tab, or ask Steward to propose them. See Relationships.

After you create a Model

Review what was generated before your AI Analysts rely on it:

1. Check the description and guidance on the Overview tab.
2. Review dimension types and expressions on the Dimensions tab, and measure expressions on the Measures tab.
3. Add filters and relationships.
4. Add Tags if your AI Analysts get their Models by Tag.
5. Give an AI Analyst access to the Model. See Models for an AI Analyst.