azure-ai-ml

What it is and what it does

Azure ML is the official Python SDK v2 for Azure Machine Learning, providing a unified client interface (MLClient) to create, manage, and execute machine learning workloads on Azure. It abstracts Azure ML resources—jobs, models, components, compute clusters, datastores—into Python objects that can be created programmatically or loaded from YAML, validated against schemas, and submitted to the service via authenticated REST calls.

The SDK supports standalone jobs (command, sweep), reusable pipeline components, AutoML for classification/regression/forecasting/vision/NLP tasks, managed online and batch inference endpoints, and resource lifecycle management. It depends on azure-core, azure-mgmt-core, azure-storage-blob, and related Azure libraries for authentication, HTTP communication, and data access. Telemetry is collected in Jupyter environments by default but can be disabled.

Use it for:

• Submit and monitor standalone training jobs (Python, R, shell scripts) to Azure compute without local execution
• Build reusable ML pipeline components and orchestrate multi-step workflows with parameter sweeps
• Run AutoML experiments for tabular classification, regression, time-series forecasting, or vision/NLP tasks
• Deploy trained models to managed online endpoints for real-time inference or batch endpoints for bulk scoring
• Manage Azure ML workspace resources (compute clusters, datastores, environments) programmatically

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

Provides a Python client library for interacting with Azure Machine Learning services, enabling job submission, pipeline orchestration, model management, and AutoML training across multiple task types.

Yes. The package is production-stable (v2 GA), actively maintained, has no known vulnerabilities, and low install friction. Install it if you need to programmatically interact with Azure Machine Learning services—it is the official and recommended SDK for that purpose. Verify your Azure subscription and workspace setup before use.

Install

azure-ai-ml on PyPI

pip install azure-ai-ml

uv add azure-ai-ml

poetry add azure-ai-ml

Installing azure-ai-ml

Before you install

Low install friction with a pure-wheel distribution. Actively maintained with a release 31 days ago and 5588 repository stars. Supports Python 3.9 through 3.14 and depends on 16 runtime packages, mostly within the Azure SDK ecosystem.

License in practice

MIT license permits commercial and private use with minimal restrictions, making it suitable for most production environments.

Quickstart

pip install azure-ai-ml

from azure.ai.ml import MLClient

ml_client = MLClient(
    credential,
    subscription_id="your-subscription-id",
    resource_group_name="your-resource-group",
    workspace_name="your-workspace"
)

job = ml_client.jobs.get("job-name")

Requires an active Azure subscription, an Azure Machine Learning Workspace, and valid Azure credentials configured in your environment.

Verify before relying

• Whether telemetry collection in Jupyter notebooks can be fully disabled or only opted out at client initialization
• Performance characteristics when managing large numbers of jobs or components in a single pipeline
• Whether the package supports disconnected or air-gapped Azure environments
• What authentication mechanisms are supported beyond the example shown in the description

Package facts

Evidence: azure_ai_ml-1.34.1-py3-none-any.whl

Keywords: azure, azure sdk

Tags

More Artificial Intelligence packages