Microsoft Introduces 'Microsoft-Decision-1': Fast Decision-Making AI Model on Foundry
Microsoft has officially introduced Microsoft-Decision-1, a fast decision-making AI model on Foundry outperforming LLMs in latency and quality on structured tas
On October 10, 2026, Microsoft officially introduced 'Microsoft-Decision-1', a new artificial intelligence model specifically designed for fast decision-making. Microsoft CEO Satya Nadella announced the release, stating that the model delivers top performance on structured decision tasks, outperforming both large language models (LLMs) and other decision models in latency and quality.

Image source: Satya Nadella / Microsoft
The model is specifically targeted at structured decision tasks requiring rapid execution and high-quality outcomes.
Performance on Structured Decisions
According to Satya Nadella, Microsoft-Decision-1 provides key advantages on structured decision tasks:
- Latency and Quality: Microsoft stated that the model outperforms both LLMs and other decision models in both latency and decision quality.
- Fast Decision Focus: The model is optimized for fast judgment and decision-making on structured tasks.
Internal Testing Across Microsoft Workflows
Microsoft revealed that it is already testing Microsoft-Decision-1 internally across a wide range of operational workflows:
- Incident Response: Tested in internal incident response workflows.
- Quality Control: Tested in internal quality control processes.
- Scientific Discovery: Applied across research and scientific discovery workflows.
Deployment and Availability: Available Now in Foundry, Coming Soon to OpenRouter
Microsoft-Decision-1 is rolling out across developer platforms:
- Available Now in Foundry: The model is immediately accessible through Microsoft Foundry.
- Coming Soon to OpenRouter: Microsoft stated that access via OpenRouter will be supported soon.
Sources
- Satya Nadella Official X (@satyanadella): Microsoft-Decision-1 Model Announcement
- Microsoft CommandLine: Microsoft-Decision-1 Model on Foundry