REDMOND, October 11 — Microsoft has introduced Microsoft-Decision-1, a new AI model designed for fast and accurate decision-making. The model, available in Microsoft Foundry and OpenRouter, outperforms both Large Language Models (LLMs) and other decision models in latency and quality on structured decision tasks. Microsoft-Decision-1 is designed for routing, classification, prioritization, verification, and workflow control, making it easier to incorporate decision intelligence into existing applications, agents, and workflows.
The model achieved the highest accuracy in a 36-benchmark comparison, spanning nearly 150,000 questions across benchmarks kept blind from training. It was also the fastest measured, 2.5 times quicker than H2O-Lightning-4B v1.1 and 35 times quicker than GPT-6 Sol. Microsoft-Decision-1 provides a calibrated probability score for each option, supporting yes/no, multiple-choice, and rating options, as well as rubric-based grading of AI responses and agent actions.
The model was evaluated across dozens of benchmarks, spanning routing, ranking, long context, multilingual and out-of-distribution tasks, reasoning, and safety, demonstrating strong generalization. In production, Microsoft-Decision-1 changes its decision on 1.3% of perturbations on average, with zero flips when option descriptions are paraphrased or when options are reversed or shuffled. The model is expected to be rebased on other models, including Microsoft AI (MAI) and OpenAI in the future.
Microsoft Azure, the cloud computing platform that hosts Microsoft Foundry, was first introduced at the Professional Developers Conference in October 2008 under the code name “Project Red Dog”. It was officially launched as Windows Azure in February 2010 and later renamed to Microsoft Azure on March 25, 2014. The platform offers more than 600 services and provides a service level agreement guaranteeing 99.9% availability for applications and data hosted on its infrastructure, subject to specific terms outlined in the SLA documentation.
Starting in 2022, virtual machines on Azure have been powered by Ampere Cloud-native processors, with most users running Linux distributions, including Microsoft’s own Azure Sphere. Azure Kubernetes Service provides the capability to deploy production-ready Kubernetes clusters in the cloud, supporting scalable AI workloads like Microsoft-Decision-1. Entra ID, formerly Azure AD, enables secure identity management across hybrid environments, allowing organizations to synchronize on-premises directories and implement single sign-on for users accessing AI tools and services.
These foundational cloud capabilities form the backbone for deploying and managing advanced models such as Microsoft-Decision-1 at enterprise scale, ensuring reliability, security, and seamless integration into existing workflows. As the model evolves through future rebasing on Microsoft AI and OpenAI systems, its integration within Azure’s ecosystem will be critical to maintaining performance and accessibility for developers and organizations worldwide.
