New column added to Lakehouse table, not appearing in Semantic Model

[Joe_Barry]

I am building a semantic model using Direct Lake. I added a new column to a table in my Lakehouse, I see it in the SQL Endpoint, but it doesn't come through to the Semantic Model. 

In Service edit tables is greyed out and in Desktop it doesn't appear, even though I am in Edit Mode. 

After removing the table and re-adding it, the new columns appear. Please don't tell me this is the only way of doing this? I have to redo all the relationships again.

Thanking you in advance

Joe

[v-sgandrathi]

Hi @Joe_Barry,

We wanted to follow up since we haven't heard back from you regarding our last response. We hope your issue has been resolved.

If you need any further assistance, feel free to reach out.

Thank you for being a valued member of the Microsoft Fabric Community Forum!

[v-sgandrathi]

Hi @Joe_Barry,

Just wanted to check regarding your question. We haven’t heard back and want to ensure you're not stuck. If you need anything else or have updates to share, we’re here to help!

Thank you.

[v-sgandrathi]

Hi @Joe_Barry,

This issue is likely related to the version of Power BI Desktop used when the semantic model was first created. Direct Lake models built with older Desktop versions (like November 2025) use an earlier metadata contract. When you upgrade to the January 2026 build, these existing models still function but remain in a legacy state, which means new service features, such as Edit tables and schema synchronization aren’t fully supported. As a result, new columns appear in the Lakehouse SQL endpoint but don’t show up in the semantic model, and Edit tables remains unavailable even if the tenant setting is enabled.

New semantic models created in the Fabric Service or with the latest Desktop version work as expected, since they use the updated Direct Lake engine and support schema changes. This confirms the issue is not about permissions, but about model version and lineage. The older semantic model is locked to previous Direct Lake behavior and can’t use the new sync features.

The only dependable solution is to recreate the semantic model using the latest version of Desktop or directly in the Fabric Service. After rebuilding, future schema changes can be managed with Edit tables without needing to drop and re-add tables. You can copy measures, relationships, and roles from the old model to the new one using a modeling tool like Tabular Editor to minimize rework.

Thank you.

[sean_cochran]

Can you provide a link to Microsoft documentation that explains this behavior?

I have encountered this issue with direct lake models created entirely in the service as well.

[v-sgandrathi · Accepted solution]

Hi @sean_cochran,

This behavior is a known limitation of Direct Lake semantic models and the way schema changes are synchronized.

As noted in Microsoft documentation, Direct Lake semantic models do not automatically detect schema changes (such as new columns) from the Lakehouse. When the Lakehouse table structure changes, you need to manually refresh the semantic model metadata using Edit tables. If this option is unavailable or greyed out, the model cannot update to reflect the new schema.

Microsoft also states that Direct Lake features like “Edit tables” have changed over time. Semantic models created with earlier versions of Power BI Desktop or early Microsoft Fabric Direct Lake implementations may continue working but could remain in a legacy metadata state. In this state, newer features like schema synchronization are not fully supported, even if editing is enabled in the tenant settings.

This is why:

New columns show up in the Lakehouse SQL endpoint

The semantic model does not display those columns

Edit tables stays disabled

Recreating the semantic model resolves the issue

According to Microsoft’s guidance, the most reliable solution is to recreate the semantic model using the latest Fabric Service or Power

BI Desktop version. After recreating, future schema changes can be synchronized using Edit tables, so you won’t need to drop and re-add tables.

Relevant Microsoft documentation:

Direct Lake overview - Microsoft Fabric | Microsoft Learn
Power BI Semantic Models - Microsoft Fabric | Microsoft Learn
Edit tables for Direct Lake semantic models - Microsoft Fabric | Microsoft Learn
Develop Direct Lake semantic models - Microsoft Fabric | Microsoft Learn

Thank you.

[Joe_Barry]

Thanks for the answers,  @AntoineW 

@cengizhanarslan I do have Users can edit semantic models in the Service = enabled
@AntoineW  Even after selecting a table, the Edit tables is still greyed out

I just created a new semantic model in Fabric Service and Edit Tables is available, also when I edit in desktop.

I also followed the same steps I took in Desktop and now it appears for a new model.

All I can think of, is that I created the Connection to the Lakehouse and created the Semantic model in Nov. 2025 version of Desktop and somehow there are now issues with thensync after updating to Jan 2026

[cengizhanarslan]

With Direct Lake, the semantic model does not automatically pick up schema changes (new columns) from the Lakehouse table.

In the semantic model:

1. Open the semantic model in Edit mode (Service or Desktop, depending on how it was created)

2. Go to Edit tables

3. Select the Lakehouse table (even if you don’t change anything)

4. Click OK / Apply

Since you said it is not available, greyed out I believe you should enable that first by:

• Tenant setting: Users can edit semantic models in the Service = enabled

[AntoineW]

Hi @Joe_Barry,

Thanks for your question, It may seem like a problem of sync between semantic model and lakehouses.

I have also this issue before, but what I've done, in Service, by clicking on a table (any table), I could click on "Edit Tables" and confirm, that synchronizes lakehouse and semantic model : 

Of course, one of the option is to drop and recreate the table, but if you have a lot of schema changes on the source, try to sync this way.

In production try to avoid as much as possible the additionnal columns.