Properly use live semantic model created on DataLake on OneLake

Hi jaryszek,

Thank you for reaching out to the Microsoft Fabric Community Forum.

When you use “Connect to a live semantic model” in Power BI, your new report is what we call a thin report. In this mode, the report does not bring in the original storage mode (like Direct Lake). Instead, it connects to the dataset through the Analysis Services (SQL/SSAS) endpoint. That is why you are seeing SQL Server Analysis Services in the model view.

Your original semantic model in Fabric is still running in Direct Lake mode. Thin reports will always show the SSAS endpoint, but performance is still backed by Direct Lake in the dataset itself.

If your goal is to build a report that directly uses Direct Lake (instead of just connecting to the semantic model), you will need to:
1. In Power BI Desktop, use Get Data → One Lake Datahub → Lakehouse (or Warehouse).
2. Build your dataset in Direct Lake mode from there.
3. Publish this dataset + report to the Service.
That way, your report is powered directly by Direct Lake.

Refer these links:
1. https://learn.microsoft.com/en-in/fabric/fundamentals/direct-lake-overview 
2. https://learn.microsoft.com/en-in/power-bi/connect-data/service-datasets-understand 
3. https://learn.microsoft.com/en-in/power-bi/connect-data/desktop-report-lifecycle-datasets 

Hope this clears it up. Let us know if you have any doubts regarding this. We will be happy to help.

Thank you for using the Microsoft Fabric Community Forum.