Share via

Azure AI Foundry – Can I schedule automatic retraining for a custom model?

idamin i 20 Reputation points
2025-10-25T20:34:58.16+00:00

I’ve deployed a custom text classification model in Azure AI Foundry, and it works well. However, my dataset gets updated frequently, and I want the model to retrain automatically on a schedule without manual intervention.

Is there a supported way in Foundry to set up periodic retraining for custom models? If not, what’s the best workaround to automate retraining and redeployment while keeping the endpoint live?

Azure AI Document Intelligence
0 comments
Answer accepted by question author
  1. Azar 31,135 Reputation points MVP Volunteer Moderator
    2025-10-25T21:11:05.7233333+00:00

    Hi there

    Yes, you can approach this in a couple of ways. Currently, Azure AI Foundry doesn’t have a fully built-in “automatic retrain” scheduler for custom models like a cron job inside the portal. However, you can automate retraining using Azure pipelines or Azure Functions. The usual approach is to:

    Store your updated dataset in Azure Blob Storage or Data Lake.

    Trigger a pipeline or function on a schedule (daily/weekly) using Azure Logic Apps, Azure Functions Timer Trigger, or Azure DevOps pipeline.

    Call the Foundry API to start a new training job programmatically.

    Deploy the new model once training completes, either by updating the existing endpoint or creating a new endpoint version.

    This way, your endpoint can stay live while you refresh the model regularly. Keep in mind to monitor quotas and retraining costs, especially for large datasets or frequent schedules.

    1 person found this answer helpful.
    0 comments

0 additional answers

Sort by: Most helpful
Most helpful Newest Oldest

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.