Microsoft's Decision-1 decision-scoring model enters preview
What happened
On October 26, Hacker News' front page described how Microsoft's decision-scoring model, Microsoft-Decision-1, works. Post-trained from Qwen3.5-9B, it outputs calibrated probabilities for preset answer options in a single forward pass, for routing, classification, prioritization and verification. It accepts up to 32,000 tokens of text input but does not generate explanations or handle multimodal content. The model is available in public preview through Microsoft Foundry and as a hosted API on OpenRouter.
Written by AI from the coverage · updated 1 hour ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageMicrosoft Decision-1: 9B Qwen Model Launch Explained
Microsoft-Decision-1 基于 Qwen3.5-9B 后训练，以单次前向计算为预设答案选项输出校准概率，用于路由、分类、优先级排序和验证。模型支持最多 32,000 tokens 的文本输入，不生成解释或处理多模态内容，已通过 Microsoft Foundry 公共预览及 OpenRouter 提供托管 API。
Heat over time
Not enough continuous observations to draw a trend yet.