Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
The Microsoft Dataverse semantic model automatically regenerates every 12 hours. To request an earlier run after changing signal settings or Dataverse metadata, trigger a manual regeneration instead of waiting for the scheduled cycle. Regeneration is asynchronous: requesting a run doesn't make the updated model available immediately.
Trigger a manual regeneration
Open the model you want to review from the semantic models page in Power Apps. Use the environment and permissions described in Fine-tune semantic model signals.
- Save pending signal, description, or table-source changes by selecting Save. Regenerate isn't a save command.
- Select Regenerate in the model's command bar. On a narrow screen, the command can appear in the overflow menu.
- Wait while Regenerating... is displayed. The command is unavailable while a request or regeneration is in progress.
- After the run completes, review Last regeneration and check the generated context on the model's configuration page. The timestamp is shown in local time.
A message that regeneration started successfully confirms that the request was accepted, not that generation has completed. If the request fails, review the error message; don't assume the model has been updated.
Note
- Regeneration runs across all semantic models in the environment.
- Save changes before selecting Regenerate. Unsaved edits aren't included in the request.
- Last regeneration identifies the last successful run; it doesn't confirm that changes made after that run are reflected in the model.
Regenerate after enabling sample rows
When you turn on Sample table rows, a dialog can offer Save & regenerate. Select it to save the settings before requesting regeneration. If saving fails, regeneration doesn't start.
Skip for now closes the dialog without saving or starting regeneration. Select Save separately if you want to keep the changed settings, then regenerate when ready. Sample rows aren't available to agents until the model is regenerated.