Adding field parameters in Fabric Lakehouse Semantic Models

[ARUNnambiar123]

Hey all,

I had created semantic models in Fabric in lakehouse (service) itself and wanted to create field parameters in the semantic model for aiding some PowerBI dashboards. Being new to Fabric I tried the traditional way of trying to download and add a field parameter, seems like I cannot download it and there is no option to create a parameter in Service. Does anyone know how to create field parameters in Fabric Lakehouse Semantic Models?

[v-veshwara-msft]

Hi @ARUNnambiar123 ,

Thanks for reaching out to Microsoft Fabric Community.

Field parameters can't be created directly within semantic models authored in the Fabric service UI at the moment.

That said, if you're using Power BI Desktop with a PBIP project connected to a Direct Lake semantic model (remote modeling), you can now create field parameters directly from the Modeling tab. This was recently enabled and supports scenarios where you're building reports over a service-based semantic model.

Reference: Power BI June 2025 Feature Summary | Microsoft Power BI Blog | Microsoft Power BI

If you're working purely in the service and need to create field parameters, one option is to use a DAX calculated table to manually simulate them. This involves defining a table that combines the fields or measures you want users to switch between and using that in visuals along with a slicer.

Eg: 

FieldParameter = 
UNION(
    SELECTCOLUMNS(VALUES('SalesTable'[SalesAmount]), "Name", "Sales", "Value", 'SalesTable'[SalesAmount]),
    SELECTCOLUMNS(VALUES('SalesTable'[Profit]), "Name", "Profit", "Value", 'SalesTable'[Profit])
)

Use the "Name" column in a slicer and bind "Value" to your visual to allow dynamic switching between measures or fields.

Please refer to video shared by @suparnababu8 for step by step approach.

Another approach is to enable the XMLA endpoint (read-write) in your workspace and connect to the semantic model using either SQL Server Management Studio (SSMS) or Tabular Editor. Both tools allow you to edit the model directly, you can create calculated tables or use the built-in field parameter feature in Tabular Editor to define the parameter logic, which will then be available in the service model.

Hope this helps. Please reach out for further assistance.
If this post helps, then please consider to give a kudos and Accept as the solution to help the other members find it more quickly.

Thank you.

[v-veshwara-msft]

Hi @ARUNnambiar123 ,

As we haven't heard from you in a while, just checking in to see if any responses were helpful. If so, kindly consider marking the helpful reply as 'Accepted Solution' to help others with similar queries. 

Otherwise, feel free to reach out for further assistance.

Thank you.

[burakkaragoz]

Hi @ARUNnambiar123 ,

Creating field parameters in Fabric Lakehouse Semantic Models is a bit different compared to the traditional Power BI Desktop workflow.

Here’s what you need to know:

Current Limitations:

• As of now, the Fabric Service does not support downloading semantic models (datasets) created in Lakehouse directly into Power BI Desktop, which is typically required for adding field parameters the classic way.
• The "Field parameter" feature in Power BI is not yet natively available in the Service UI for semantic models created directly in Fabric Lakehouse.

Workarounds:

1. If you have access to the original PBIX file:

   • Open the PBIX in Power BI Desktop.
   • Add your field parameters via the “Modeling > New Parameter” option.
   • Publish the updated model back to Fabric.

2. If the semantic model only exists in Fabric Service (and not as a PBIX):

   • Unfortunately, you cannot currently add field parameters directly in the Service.
   • You may need to recreate the semantic model in Power BI Desktop, add your parameters, and re-publish.
   • Keep an eye on the Fabric roadmap—this feature is highly requested and may be added soon!

3. Direct Lake Mode Models:

   • For models using Direct Lake mode, Microsoft has started rolling out more modeling features over time, but field parameters are still primarily a Desktop feature.

If you need step-by-step instructions for any of these approaches, or want guidance on structuring your semantic model for future extensibility, let me know!

Hope this clarifies the current state and your options!

[suparnababu8]

Hi @ARUNnambiar123 

If you haven;t tried this, can you please check this tutiorial, it might helps you to create field parameters in Fabric Lakehouse Semantic Models

https://youtu.be/AOgbJBbvwR4?si=HmyZ4lELjbDEsNTT

Creating Field Parameter in Direct Lake mode Semantic Model in Fabric

Thank you!

Did I answer your question? Mark my post as a solution!

Proud to be a Super User!