Live edit of Direct Lake models in Power BI Desktop

Direct Lake speeds up data-driven decisions by unlocking incredible performance directly against OneLake while ensuring maximum data reusability across Fabric. Last month, we announced a major update for developers working on Direct Lake models: Live editing of Power BI semantic models in Direct Lake mode. Now, you can use Power BI Desktop to edit Direct Lake semantic models, improving your data modeling experience and allowing export to Power BI Project (PBIP) for professional development workflows.

With live edit every modification is applied to the semantic model in the workspace, ensuring a seamless and efficient workflow.

Get started

Start by enabling the preview feature, go to File > Options and settings > Options > Preview features and check the box next to “Live edit of Power BI semantic models in Direct Lake mode”

Once the preview feature is enabled, restart Power BI Desktop, open OneLake datahub:

Then, choose the Direct Lake semantic model you want to modify and click on Edit:

By opening the model for editing in Power BI Desktop, you can directly modify the semantic model. Since it’s a Live Edit, all changes are instantly applied to the semantic model in the Fabric workspace without needing to save. Changes include all modeling tasks, such as renaming tables/columns, add/remove tables from Lakehouse, creating measures, and creating calculation groups. DAX query view is available to run DAX queries to preview data and test measures before saving them to the model.

Enterprise development with Power BI Project (PBIP)

This feature also allows for the integration of Power BI Project with Direct Lake semantic models. You can export the definition of your semantic model after opening it for editing to a Power BI Project, which provides a local copy of the semantic model and report metadata, enabling source control with Git and deployment mechanisms such as Fabric Git Integration.

Go to File > Export > Power BI Project and export it as a Power BI Project:

When opening a Power BI Project with a Direct Lake semantic model, you need to choose a remote semantic model in a Fabric workspace, as direct lake semantic models cannot run locally in Power BI Desktop.

Each developer should choose their own isolated semantic model in a development workspace to prevent conflicts during development. Use Git as the central metadata repository, deploying the semantic model metadata to the final workspace via Fabric Git or Fabric APIs.