Microsoft 推出了 Decision-1，这是一款专为分类和路由等快速决策任务设计的新型 AI 模型。Decision-1 基于阿里巴巴的 Qwen3.5-9B 模型构建，在 Microsoft 的内部基准测试中，以低延迟实现了 83.5% 的准确率，性能优于其他几款模型。该模型可通过 Microsoft Foundry 和 OpenRouter 获取，为需要即时、结构化输出而非生成文本的应用提供了经济高效的解决方案。 AI
影响 加速了用于代理工作流和决策自动化的专用 AI 模型的采用。
排序理由 Microsoft AI 模型发布，附带系统卡和性能指标。
在 Mastodon — mastodon.social 阅读 →
AI 生成摘要 · Google Gemini · 来自 6 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
<p><img alt="" class="attachment-full size-full wp-post-image" height="768" src="https://the-decoder.com/wp-content/uploads/2026/06/microsoft_logo_plain.png" style="height: auto; margin-bottom: 10px;" width="1376" /></p> <p> With Decision-1, Microsoft enters the growing decision …
<p>Microsoft has released Microsoft-Decision-1, a decision model for routing, classification, verification and agent control. Microsoft-Decision-1 is a decision-scoring model that returns a calibrated probability for each fixed answer option instead of generated text. It is post-…
<h2> TL;DR </h2> <ul> <li>Microsoft-Decision-1 is a small language model that returns a probability score for each option in a list you hand it, so a program can route a request, grade a response or decide what an agent does next.</li> <li>Microsoft published the model on 9 Octob…
🧠 # Microsoft -Decision-1 porta la direzione di # Jev in Foundry: probabilità su opzioni predefinite, per microdecisioni rapide nei workflow. Anche le Decisions API di # OpenAI seguono questa strada. 👉 I dettagli: https:// lnkd.in/p/d8HUed7N ___ ✉️ 𝗦𝗲 𝘃𝘂𝗼𝗶 𝗿𝗶𝗺𝗮𝗻𝗲𝗿𝗲 𝗮𝗴𝗴𝗶𝗼𝗿𝗻𝗮𝘁𝗼/𝗮 𝘀…
Microsoft has released Microsoft-Decision-1, a reasoning model engineered for autonomous workflows and decision benchmarks. Rather than relying purely on internal architecture or OpenAI models, Microsoft constructed the initial version on Alibaba's open-weight Qwen3.5-9B base wei…
Microsoft has released Microsoft-Decision-1, a decision-scoring model built on Qwen3.5-9B for routing, classification, verification and agent control. Available via Microsoft Foundry and OpenRouter, it achieves 83.5% average accuracy across 36 benchmarks at 85ms latency. https://…