微软发布决策模型Microsoft-Decision-1
Microsoft-Decision-1, our model for fast decision-making
微软正式发布的专用决策模型，性能指标明确且大幅超越现有方案，对优化Agent路由和结构化任务有直接参考价值。
Deep Dive
深入探讨
Share
分享
in r x f
在 r x f 中
Introducing Microsoft-Decision-1, our model for fast decision-making
介绍 Microsoft-Decision-1，我们用于快速决策的模型
Introducing Microsoft-Decision-1, our model for fast decision-making
介绍 Microsoft-Decision-1，我们用于快速决策的模型
Our new decision-scoring model delivers top performance in latency and quality on structured decision tasks to outperform both LLMs and other decision models.
我们的新决策评分模型在结构化决策任务的延迟和质量方面表现出色，超越了大型语言模型（LLM）和其他决策模型。
By Achint SrivastavaVP of Software Engineering, Office of the CTO, Microsoft
作者：Achint Srivastava，微软首席技术官办公室软件工程副总裁
2026.10.09
Decision models are quickly emerging as an important new category in AI. Unlike LLMs, which are designed to generate text or reason through complex problems, decision models are purpose-built to deliver structured outputs that software can immediately act on. And once you understand that capability—making decisions and classifying things at very low cost with high performance—all kinds of useful tasks get unlocked.
决策模型正迅速成为人工智能领域的一个重要新类别。与旨在生成文本或推理复杂问题的大型语言模型（LLM）不同，决策模型专为提供软件可立即执行的结构化输出而设计。一旦你理解了这种能力——以极低的成本和高性能进行决策和分类——各种有用的任务便得以解锁。
Today we’re introducing Microsoft-Decision-1, our new model for fast decision-scoring, available in Microsoft Foundry and through OpenRouter. This model is designed for routing, classification, prioritization, verification, and workflow control, making it easier to incorporate decision intelligence into existing applications, agents, and workflows in a secure, trusted environment. Microsoft-Decision-1 delivers top performance in latency and quality on structured decision tasks to outperform both LLMs and other decision models.
今天，我们推出了 Microsoft-Decision-1，这是一款用于快速决策评分的新模型，可在 Microsoft Foundry 和 OpenRouter 上使用。该模型专为路由、分类、优先级排序、验证和工作流控制而设计，有助于在安全、可信的环境中更轻松地将决策智能集成到现有应用程序、代理和工作流中。Microsoft-Decision-1 在结构化决策任务的延迟和质量方面表现出色，超越了大型语言模型（LLM）和其他决策模型。
Microsoft-Decision-1 achieved the highest accuracy in our 36-benchmark comparison, spanning nearly 150,000 questions across benchmarks kept blind from training. And in our benchmarking, it was the fastest measured: 2.5 times quicker than H2O-Lightning-4B v1.1, the runner-up, and 35 times quicker than GPT-6 Sol.
在我们涵盖近 150,000 道问题的 36 项基准测试对比中，Microsoft-Decision-1 取得了最高的准确率，这些基准测试均对训练数据保持盲测。在我们的基准测试中，它的速度最快：比第二名 H2O-Lightning-4B v1.1 快 2.5 倍，比 GPT-6 Sol 快 35 倍。
Editor’s note: This post was updated from the original to add benchmarks for Jev on accuracy and calibration.
编辑注：本文已更新，增加了 Jev 在准确性和校准方面的基准测试结果。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力