Multiple semantic models

[cheid_4838]

I published two seperate dashboards that are linked to a semantic model.  When I publish the dashboards I get two semantic models for each version of the dashboard I published.  Why are two additional semantic models showing up for each dashboard versus being linked to the main semantic model?  I've never had this happen before.  Thanks for your help.

[v-kpoloju-msft · Accepted solution]

Hi @cheid_4838,
Thank you for reaching out to the Microsoft fabric community forum. Thank you @Deku, @danextian, for your inputs on this issue.

After reviewing the details you provided, I have identified few workarounds that may help resolve the issue. Please follow these steps:

Before addressing the solution, it's crucial to understand the root causes to prevent recurrence:

Import Mode Instead of Live Connection: Reports using import mode include their own data models, creating new semantic models upon publishing. Not Using Power BI Dataset as Data Source: Direct connections to original sources instead of a shared Power BI dataset result in a new embedded model.

Publishing to Different Workspaces: Separate workspaces without a shared dataset connection cause Power BI to treat each report as needing its own model. Re-uploading Reports Without Live Binding: Republishing edited PBIX files without a consistent dataset ID leads to duplicated semantic models.

Verify Data Source in Desktop: Open both PBIX files. Go to Model View or Data Source Settings and confirm if the report is importing data from original sources or using a live connection to an existing Power BI Dataset.

Use a Shared Semantic Model (Live Connection): In Power BI Desktop, start a new report. Use "Get Data" → "Power BI Datasets" → select the existing dataset from the Power BI Service. This connects your report live to the existing semantic model, avoiding the creation of a new model upon publishing.

Republish Reports: After connecting your report to the shared dataset, republish it. Power BI will only publish the report, preventing dataset duplication. In Power BI Service, go to the workspace and delete the extra semantic models that were mistakenly created.

Maintain a centralized dataset strategy: Publish one master semantic model and have all reports connect to it. Limit who can publish datasets and train report authors to use live connections.

Kindly refer to the below links for more information:
Connect to semantic models in Power BI - Power BI | Microsoft Learn
Introduction to semantic models across workspaces - Power BI | Microsoft Learn

If this post helps, then please give us ‘Kudos’ and consider Accept it as a solution to help the other members find it more quickly.

Thank you for using Microsoft Community Forum.

[v-kpoloju-msft]

Hi @cheid_4838,

May I ask if you have resolved this issue? If so, please mark the helpful reply and accept it as the solution. This will be helpful for other community members who have similar problems to solve it faster.

Thank you.

[danextian]

Hi @cheid_4838 

Connecting to a semantic model using DirectQuery creates a local model that's still linked to the original. If you want to avoid this, consider using a live connection to build thin reports instead

[Deku]

This is likely because you have a composite model where one of your groups is a a Semantic model connected vai direct query over analysis services