Microsoft Ships Decision-1, a Qwen3.5-9B Classifier Model Priced at $0.042/M Input and Free Output

Microsoft launched Decision-1, a specialized model for routing, classification and rubric-based evaluation that returns calibrated probabilities on yes/no, multiple-choice and rating-scale tasks. Microsoft claims highest accuracy across 36 benchmarks and nearly 150,000 questions, 4.5x faster than Quyet-1.0-Large and 35x faster than GPT-6 Sol, with decisions perturbation-stable 98.7% of the time. Pricing is $0.042 per million input tokens with output free; the model is based on Qwen3.5-9B and available via Microsoft Foundry and OpenRouter.