Microsoft has released Microsoft-Decision-1, a small model built for a narrow job: given a situation, a question and a fixed set of options, it returns a calibrated probability for each option. It is meant for routing, classification, prioritisation, verification and workflow control inside apps and agents.

The model was post-trained from Qwen3.5-9B for single-pass scoring and supports yes/no, multiple-choice, rating-scale and rubric-style grading through API calls. Microsoft says it tested the model across 36 benchmarks covering nearly 150,000 questions, and reports much lower latency than using a frontier model for the same scoring work.

Microsoft-Decision-1 is available through Microsoft Foundry and OpenRouter, priced at $0.042 per million input tokens in the launch reporting. It is not a chat model: Microsoft is explicit that it is not designed for open-ended generation, conversation or summarisation.