Full guide: What is a decision model? AI decision models explained
A decision model reads a state, the content you hand it, and answers typed questions about it. Each answer is one of the options you defined, with a probability for every option, so your code can branch on it. The model writes no prose.
Cloudflare, Microsoft, Liquid AI and Nace.AI all call their models decision models. TypeSafe AI, whose Jev was the first, calls the wider group System One models. Since 19 October 2026 that group has a second class, extraction models, which return typed values read from the input instead of decisions.
A concrete example: a support ticket goes in as state, and a Choice question with the options shipping, returns and billing comes back as {"choice":"returns","confidence":1.0,"probabilities":{"shipping":0.0,"returns":1.0,"billing":0.0}}. The router switches on the string.
The decision model guide covers the question types, the vendors’ names and how it differs from an LLM and a classifier.