We’re proud to share that Microsoft has once again been named a Leader in the 2025 Gartner® Magic Quadrant™ for Data Science and Machine Learning (DSML) Platforms.
We’re proud to share that Microsoft has once again been named a Leader in the 2025 Gartner® Magic Quadrant™ for Data Science and Machine Learning (DSML) Platforms. We believe this recognition reflects our continued commitment to providing organizations with a comprehensive toolchain for building and deploying machine learning models and AI applications, transforming how businesses operate. Azure Machine Learning is part of a broad, interoperable ecosystem across Microsoft Fabric, Microsoft Purview, and within Azure AI Foundry.
Gartner defines a data science and machine learning platform as an integrated set of code-based libraries and low-code tooling. These platforms support the independent use and collaboration among data scientists and their business and IT counterparts, with automation and AI assistance through all stages of the data science life cycle, including business understanding, data access and preparation, model creation, and sharing of insights. They also support engineering workflows, including the creation of data, feature, deployment, and testing pipelines. The platforms are provided via desktop client or browser with supporting compute instances or as a fully managed cloud offering.
Leading the way in 2025
With Microsoft, we’re turning our media expertise into a competitive advantage—and harnessing data to build brands and drive business growth.
—Callum Anderson, Global Director for DevOps and SRE at Dentsu.
At Microsoft, we envision a unified experience where data scientists, AI engineers, developers, IT operations professionals, and business users come together to create applications and manage the entire AI lifecycle across personas and projects. To that end, in November 2024, we announced the availability of Azure AI Foundry—a platform that allows developers to design, customize, and manage AI applications. Azure Machine Learning is a trusted workbench that exists on top of Azure AI Foundry and powers the underlying tool chain technology, with capabilities for model customization, including fine-tuning and RAG.
Advancing AI with Azure Machine Learning and intelligent agents
As part of Azure AI Foundry, the Foundry Agent Service empowers developer teams to orchestrate AI agents that automate complex, cross-functional workflows. Whether building solutions for software engineering, business process automation, customer support, or data analysis, Foundry Agent Service provides a robust, secure, and interoperable foundation to operationalize AI agents in production environments.
- With support for multi-agent orchestration, developers can design agent systems that coordinate across tasks, share state, recover from failures, and evolve flexibly as requirements change. These agents can be grounded in enterprise knowledge using Microsoft Fabric, Bing, and SharePoint, while interacting with both proprietary and third-party tools thanks to open standards like MCP (Model Context Protocol) and A2A (Agent2Agent).
- Developers can start building locally using open-source frameworks like Semantic Kernel and AutoGen, and we’re on a clear path toward delivering a unified SDK across the two frameworks and Azure AI Foundry that allows you to move from local experimentation to production in cloud without rewriting any code. This ensures consistent developer experience—from initial prototyping to managed orchestration with observability and enterprise-grade control.
Together, Azure Machine Learning and Foundry Agent Service enable a future where AI systems are designed for enterprise use with scalability and security in mind.
Leveraging AI models with Azure AI Foundry
Azure AI Foundry offers developers an innovative method of deploying and managing its over 11,000 AI models with tools like the Model Router, Model Leaderboard, and Model Benchmarks.
- The Model Leaderboard simplifies the comparison of model performance across real-world tasks, providing transparent benchmark scores, task-specific rankings, and live updates, enabling users to select the high accuracy, fast throughput, or competitive price-performance ratio efficiently.
- Model Benchmarks in Azure AI Foundry offer a streamlined way to compare model performance using standardized datasets, while also allowing customers to evaluate models on their own data to identify the best fit for their specific scenarios.
- Complementing this, the Model Router—available now for Azure OpenAI models—dynamically routes queries to the most suitable large language model (LLM) by assessing factors such as query complexity, cost, and performance, ensuring high-quality results while minimizing compute expenses.
These capabilities empower businesses to deploy flexible and adaptive AI systems with enterprise-grade performance, security, and governance. With integrated innovation from Microsoft and its ecosystem, users gain access to future-ready solutions that enhance efficiency and scalability, ensuring they stay ahead in the rapidly evolving AI landscape.
Optimizing AI performance with fine-tuning in Azure AI Foundry
Fine-tuning is an essential tool for organizations aiming to customize pre-trained AI models for specific tasks, enhancing their performance, accuracy, and adaptability, all while reducing operational costs. Fine-tuning in Azure AI Foundry is powered by the underlying Azure Machine Learning tool chain.
- With innovations such as Reinforcement Fine-Tuning (RFT) using the o4-mini model, Azure AI Foundry enables developers to improve reasoning, context-aware responses, and dynamic decision-making through reinforcement signals. This adaptability is particularly suited for applications requiring ongoing learning, making it an ideal method for evolving business logic and ensuring models stay relevant in dynamic environments.
- Azure AI Foundry further simplifies fine-tuning with features such as Global Training and the Developer Tier. Global Training lowers costs by allowing model customization across multiple Azure regions, giving developers flexibility and scalability while adhering to strict privacy policies. The Developer Tier offers an affordable way to evaluate fine-tuned models, enabling simultaneous testing across deployments and empowering users to choose the best candidate for production with precision and efficiency.
Together, these capabilities enable developers and enterprises to unlock the full potential of their AI systems, driving innovation and efficiency in the rapidly evolving digital landscape.
Enabling organizations to deploy AI solutions
From healthcare and finance to manufacturing and retail, customers are using Azure Machine Learning to solve complex problems, optimize operations, and unlock new business models. Whether it’s deploying foundation models, orchestrating AI agents, or scaling real-time inference, Microsoft is helping organizations turn data into impact.
Begin your journey with Azure Machine Learning
The migration to Azure is just the beginning. We’ve laid the foundation to explore opportunities we could only imagine before.
—Steve Fortune, Chief Digital and Technology Officer at CSX.
Machine learning is revolutionizing the operational and competitive landscape for businesses in the digital age. It offers opportunities to optimize business processes, improve customer experiences, and drive innovation. Azure Machine Learning serves as a robust and versatile platform for machine learning and data science, enabling organizations to implement AI solutions responsibly and effectively.
Gartner, Magic Quadrant for Data Science and Machine Learning Platforms, By Afraz Jaffri, Maryam Hassanlou, Tong Zhang, Deepak Seth, Yogesh Bhatt, 28 May 2025.
GARTNER is a registered trademark and service mark of Gartner, Inc. and/or its affiliates in the U.S. and internationally, Magic Quadrant is a registered trademark of Gartner, Inc. and/or its affiliates and is used herein with permission. All rights reserved.
This graphic was published by Gartner, Inc. as part of a larger research document and should be evaluated in the context of the entire document. The Gartner document is available upon request from [https://www.gartner.com/en/documents/6533902].
Gartner does not endorse any vendor, product or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings or other designation. Gartner research publications consist of the opinions of Gartner’s research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.