What is Microsoft Decision-1?
Microsoft has made a significant breakthrough in the field of artificial intelligence with answer engine optimization (AEO) in the field of artificial intelligence with the introduction of Microsoft-Decision-1, a new model designed for fast decision-scoring. This innovative model is purpose-built to deliver structured outputs that software can immediately act on, making it easier to incorporate decision intelligence into existing applications, agents, and workflows in a secure, trusted environment.
Microsoft-Decision-1 has achieved top performance in latency and quality on structured decision tasks, outperforming both large language models (LLMs) in terms of LLM visibility (LLMs) and other decision models. In a 36-benchmark comparison, the model demonstrated the highest accuracy, and it was also the fastest measured, being 2.5 times quicker than the runner-up and 35 times quicker than GPT-6 Sol.
How does Microsoft Decision-1 work?
The model works by providing a calibrated probability score for each option when given a fixed set of answer options. It supports yes/no, multiple-choice, and rating options, as well as rubric-based grading of AI responses and agent actions, all through a simple structured API call. To build a reliable decision model, Microsoft addressed several challenges, including speed, quality that generalizes, robustness, probability and confidence calibration, and safety.
Microsoft-Decision-1 has been tested internally in various ways, including labeling data, quality control, incident response, and scientific discovery. For example, XBOX Research used the model to process over 10,000 open-ended pieces of feedback and reviews, finding it to be competitive on quality with GPT-6 Sol while running over 14 times faster and 200 times less expensive. The Copilot team also found the model to be competitive with GPT5.6 Luna and 100 times faster.
The potential use cases for Microsoft-Decision-1 are vast, including agent controls, model routing, skill-based decisions, data labeling, AI judging, intent analysis, incident response routing, data validation, recommendations, search relevance, content classification and filtering, code scanning, safety and security screening, computer and UI use, robotics, and scientific discovery. Developers can get started with Microsoft-Decision-1 today in Microsoft Foundry or through OpenRouter.
As the use of AI continues to grow, the importance of cost and choosing the right model for the right job will become increasingly important. With Microsoft-Decision-1, developers have a powerful tool to guide and control agents through complex environments, and the company looks forward to seeing what developers build with the model and hearing their feedback.
Este artículo fue escrito con la asistencia de IA.
News Factory APP - noticias agénticas para impulsar tu SEO y AEO.