Interface DecisionModel

All Known Implementing Classes:
TypeSafeDecisionModel

@Experimental public interface DecisionModel
A model that evaluates an input against a set of named, typed questions and returns one typed answer per question, instead of generated text.

Typical uses are classification, routing, gating and grading: for example, deciding which team should handle a support ticket, whether a message is spam, or how urgent an incident is. All questions of a request are answered against the same input in a single call.

The supported question types are YesNoQuestion (a yes/no judgment answered with a probability), ChoiceQuestion (one option out of a named set) and ScaleQuestion (a position on an ordered scale). Implementations may support additional question types.

Probabilities in the answers must be derived from the model's output distribution, since thresholds and escalation logic in applications rely on them. An implementation that cannot provide them for choice or scale answers leaves probabilities() empty instead of inventing values.

Since:
1.21.0