Microsoft puts a 9B decision model in Foundry for agent workflows

Microsoft released Decision-1, a model that scores fixed choices instead of generating text, in Microsoft Foundry.

The model accepts text inputs of up to 32,768 tokens and returns JSON, supporting yes-or-no questions, multiple choice, ratings, rubric grading, relevance judgments and an explicit "cannot tell" option.

Microsoft reports the highest accuracy in its comparison of 36 benchmarks and nearly 150,000 questions, with input tokens priced at $0.042 per million and output tokens free.