We’re proud to share that Microsoft has as soon as once more been named a Chief within the 2025 Gartner® Magic Quadrant™ for Information Science and Machine Studying (DSML) Platforms.
We’re proud to share that Microsoft has as soon as once more been named a Chief within the 2025 Gartner® Magic Quadrant™ for Information Science and Machine Studying (DSML) Platforms. We imagine this recognition displays our continued dedication to offering organizations with a complete toolchain for constructing and deploying machine studying fashions and AI functions, reworking how companies function. Azure Machine Studying is a part of a broad, interoperable ecosystem throughout Microsoft Cloth, Microsoft Purview, and inside Azure AI Foundry.
Gartner defines an information science and machine studying platform as an built-in set of code-based libraries and low-code tooling. These platforms assist the impartial use and collaboration amongst information scientists and their enterprise and IT counterparts, with automation and AI help by all phases of the information science life cycle, together with enterprise understanding, information entry and preparation, mannequin creation, and sharing of insights. Additionally they assist engineering workflows, together with the creation of information, function, deployment, and testing pipelines. The platforms are offered through desktop consumer or browser with supporting compute situations or as a totally managed cloud providing.
Main the best way in 2025
With Microsoft, we’re turning our media experience right into a aggressive benefit—and harnessing information to construct manufacturers and drive enterprise progress.
—Callum Anderson, International Director for DevOps and SRE at Dentsu.
At Microsoft, we envision a unified expertise the place information scientists, AI engineers, builders, IT operations professionals, and enterprise customers come collectively to create functions and handle all the AI lifecycle throughout personas and initiatives. To that finish, in November 2024, we introduced the supply of Azure AI Foundry—a platform that permits builders to design, customise, and handle AI functions. Azure Machine Studying is a trusted workbench that exists on high of Azure AI Foundry and powers the underlying instrument chain know-how, with capabilities for mannequin customization, together with fine-tuning and RAG.
Advancing AI with Azure Machine Studying and clever brokers
As a part of Azure AI Foundry, the Foundry Agent Service empowers developer groups to orchestrate AI brokers that automate advanced, cross-functional workflows. Whether or not constructing options for software program engineering, enterprise course of automation, buyer assist, or information evaluation, Foundry Agent Service supplies a strong, safe, and interoperable basis to operationalize AI brokers in manufacturing environments.
- With assist for multi-agent orchestration, builders can design agent programs that coordinate throughout duties, share state, get well from failures, and evolve flexibly as necessities change. These brokers could be grounded in enterprise information utilizing Microsoft Cloth, Bing, and SharePoint, whereas interacting with each proprietary and third-party instruments due to open requirements like MCP (Mannequin Context Protocol) and A2A (Agent2Agent).
- Builders can begin constructing regionally utilizing open-source frameworks like Semantic Kernel and AutoGen, and we’re on a transparent path towards delivering a unified SDK throughout the 2 frameworks and Azure AI Foundry that permits you to transfer from native experimentation to manufacturing in cloud with out rewriting any code. This ensures constant developer expertise—from preliminary prototyping to managed orchestration with observability and enterprise-grade management.
Collectively, Azure Machine Studying and Foundry Agent Service allow a future the place AI programs are designed for enterprise use with scalability and safety in thoughts.
Leveraging AI fashions with Azure AI Foundry
Azure AI Foundry provides builders an progressive methodology of deploying and managing its over 11,000 AI fashions with instruments just like the Mannequin Router, Mannequin Leaderboard, and Mannequin Benchmarks.
- The Mannequin Leaderboard simplifies the comparability of mannequin efficiency throughout real-world duties, offering clear benchmark scores, task-specific rankings, and reside updates, enabling customers to pick out the excessive accuracy, quick throughput, or aggressive price-performance ratio effectively.
- Mannequin Benchmarks in Azure AI Foundry supply a streamlined method to examine mannequin efficiency utilizing standardized datasets, whereas additionally permitting clients to guage fashions on their very own information to determine the most effective match for his or her particular eventualities.
- Complementing this, the Mannequin Router—accessible now for Azure OpenAI fashions—dynamically routes queries to essentially the most appropriate massive language mannequin (LLM) by assessing elements corresponding to question complexity, price, and efficiency, making certain high-quality outcomes whereas minimizing compute bills.
These capabilities empower companies to deploy versatile and adaptive AI programs with enterprise-grade efficiency, safety, and governance. With built-in innovation from Microsoft and its ecosystem, customers acquire entry to future-ready options that improve effectivity and scalability, making certain they keep forward within the quickly evolving AI panorama.
Optimizing AI efficiency with fine-tuning in Azure AI Foundry
Advantageous-tuning is a necessary instrument for organizations aiming to customise pre-trained AI fashions for particular duties, enhancing their efficiency, accuracy, and flexibility, all whereas lowering operational prices. Advantageous-tuning in Azure AI Foundry is powered by the underlying Azure Machine Studying instrument chain.
- With improvements corresponding to Reinforcement Advantageous-Tuning (RFT) utilizing the o4-mini mannequin, Azure AI Foundry allows builders to enhance reasoning, context-aware responses, and dynamic decision-making by reinforcement indicators. This adaptability is especially fitted to functions requiring ongoing studying, making it a great methodology for evolving enterprise logic and making certain fashions keep related in dynamic environments.
- Azure AI Foundry additional simplifies fine-tuning with options corresponding to International Coaching and the Developer Tier. International Coaching lowers prices by permitting mannequin customization throughout a number of Azure areas, giving builders flexibility and scalability whereas adhering to strict privateness insurance policies. The Developer Tier provides an inexpensive method to consider fine-tuned fashions, enabling simultaneous testing throughout deployments and empowering customers to decide on the most effective candidate for manufacturing with precision and effectivity.
Collectively, these capabilities allow builders and enterprises to unlock the total potential of their AI programs, driving innovation and effectivity within the quickly evolving digital panorama.
Enabling organizations to deploy AI options
From healthcare and finance to manufacturing and retail, clients are utilizing Azure Machine Studying to resolve advanced issues, optimize operations, and unlock new enterprise fashions. Whether or not it’s deploying basis fashions, orchestrating AI brokers, or scaling real-time inference, Microsoft helps organizations flip information into impression.
Start your journey with Azure Machine Studying
The migration to Azure is only the start. We’ve laid the inspiration to discover alternatives we may solely think about earlier than.
—Steve Fortune, Chief Digital and Expertise Officer at CSX.
Machine studying is revolutionizing the operational and aggressive panorama for companies within the digital age. It provides alternatives to optimize enterprise processes, enhance buyer experiences, and drive innovation. Azure Machine Studying serves as a strong and versatile platform for machine studying and information science, enabling organizations to implement AI options responsibly and successfully.
Gartner, Magic Quadrant for Information Science and Machine Studying Platforms, By Afraz Jaffri, Maryam Hassanlou, Tong Zhang, Deepak Seth, Yogesh Bhatt, 28 Could 2025.
GARTNER is a registered trademark and repair mark of Gartner, Inc. and/or its associates within the U.S. and internationally, Magic Quadrant is a registered trademark of Gartner, Inc. and/or its associates and is used herein with permission. All rights reserved.
This graphic was revealed by Gartner, Inc. as half of a bigger analysis doc and must be evaluated within the context of all the doc. The Gartner doc is obtainable upon request from [https://www.gartner.com/en/documents/6533902].
Gartner doesn’t endorse any vendor, services or products depicted in its analysis publications, and doesn’t advise know-how customers to pick out solely these distributors with the very best rankings or different designation. Gartner analysis publications encompass the opinions of Gartner’s analysis group and shouldn’t be construed as statements of truth. Gartner disclaims all warranties, expressed or implied, with respect to this analysis, together with any warranties of merchantability or health for a selected objective.