Microsoft-Decision-1 is a decision-scoring model designed for routing, classification, verification, and agent control. Given a fixed set of options, it returns calibrated probabilities for each in a single pass as structured output that software can act on directly. It is post-trained from Qwen3.5-9B and, per Microsoft, is about 35x faster than GPT-6 Sol at median latency. It is available in Microsoft Foundry at $0.042 per million input tokens, with free output.
Microsoft-Decision-1
Microsoft’s decision model for agents and workflows
Microsoft-Decision-1 Introduction
Key Features
- Decision-scoring model for routing, classification, verification, and agent control
- Returns calibrated probabilities for a fixed set of options in a single pass
- Outputs structured results that software can act on directly
- Post-trained from Qwen3.5-9B
- Per Microsoft, about 35x faster than GPT-6 Sol at median latency
- Available in Microsoft Foundry
Use Cases
- Routing requests among fixed options in software workflows
- Classification tasks requiring calibrated probability scores
- Verification steps in automated pipelines
- Decision control for AI agent systems
Why Startups Use It
Startups building agentic or automated workflows can use a specialized decision-scoring model to get fast, structured, calibrated outputs for routing, classification, and verification, with per-token pricing in Microsoft Foundry.
Alternative Options
GPT-6 Sol
Frequently Asked Questions
What is Microsoft-Decision-1?
It is a decision-scoring model that assigns calibrated probabilities to a fixed set of options in a single pass, producing structured output for software to act on.
What is it built on?
It is post-trained from Qwen3.5-9B.
Where is it available and what does it cost?
It is available in Microsoft Foundry at $0.042 per million input tokens, with free output.
More About Microsoft-Decision-1
PricingPaid
Listed DateOct 11, 2026
Authority Badge
Add our badge to your website to showcase product credibility and listing status.
Featured List