The category name comes from Daniel Kahneman’s Thinking, Fast and Slow, which splits thinking into fast, intuitive System 1 and slow, deliberate System 2 reasoning. A System One model takes the fast half: a judgment your code needs in milliseconds, returned in a shape your code can branch on without parsing prose.
TypeSafe’s docs define three question types. A Choice picks one option from a set of up to 255 and returns a probability across all of them. A Score rates content against 2 to 10 ordered levels. A Noul returns a single probability that the answer to a yes/no question is yes.
One worked example from the docs: a bug report saying the export button crashes the settings page in Safari but works in Chrome. Scored against three severity levels, the answer is 1.3 with confidence 0.54 and level probabilities of 0.0, 0.7 and 0.3. The 1.3 is the probability-weighted mean of the level numbers. The 0.3 sitting on the worst level is the part a triage rule would act on.
The model never picks the next action. Code owns the control flow and any side effects, which is the split described in how to build with System One models.