Evaluation model vs chat model

A chat model returns text you have to trust and parse. An evaluation model returns a probability per option, which is what lets a simulation put a real threshold on a decision instead of a vibe.

Side by side

Chat modelEvaluation model
Outputtokensa distribution per question
Nine questionsnine calls, or one call and a parseone call
Confidenceself-reported, if askedthe mass on the chosen answer
Refusals / drifta real failure modeno free text to drift
Bills oninput and outputinput only, here

The confidence difference is the whole thing

Ask a chat model how sure it is and you get a number it wrote about itself, which is a different object from the probability it assigned. Ask an evaluation model the same question and the answer is already a probability: a character's confidence is the mass on the move they picked.

That makes a rule like Nyx ignores her own reads below 0.72 a real bar. Against a measured distribution (median P(top choice) ≈ 0.57, p75 ≈ 0.77) 0.72 is roughly her top quartile, so she overrides herself often enough to be a character and rarely enough to still act.

Where chat models are still the right call

Anything whose output is text a human reads: dialogue, summary, narration. Parlour has none of that on purpose. The ticker lines are templates filled by the engine, so the room is never describing itself in a voice it might lose.

Read next

More on the model