What are the differences between a pbix file and a pbit file?

Separating Power BI files into datasets and visuals is a good practice for several reasons:

It helps to keep the data model and report development separate, which is especially useful when data modelers and report authors are different people or teams.

It allows for the creation of multiple reports from a single data model, reducing duplication of effort and creating a more sustainable approach to maintaining your assets - a change to a metric definition would be made in one semantic model only and automatically appear in all reports

Simplifies version control - allowing you to track changes to the semantic model separately to visualisation changes.

Simplifies workspace management - giving access to the dataset no longer exposes visualisation at the same time.

It helps to preserve the model interface, which can be broken if changes are made to the model.

It can help to reduce the amount of work required to create and maintain measures and row-level security.

Large datasets (in particular those that leverage incremental refresh, which means reloading your dataset can be timely, and costly) can be updated independantly to the visuals. For instance, if you have a large dataset that is used by multiple reports, it is easier to manage the dataset in a separate file. This way, you can make changes to the dataset without affecting the reports that use it. Similarly, if you have a report that uses multiple datasets, it is easier to manage the report in a separate file. This way, you can make changes to the report without affecting the datasets that it uses.

---