A team uses a general-purpose Foundry model to classify incoming emails as either urgent or not urgent and feed that label into an automated routing system. During testing, the same email sometimes gets classified differently on repeated runs, which breaks the routing logic downstream. Without changing the model or retraining anything, what should the team adjust to make the classification output more consistent across repeated runs?

Lowering the temperature parameter toward zero is the correct adjustment because temperature controls how much randomness the model introduces when choosing among likely next words; lowering it toward zero makes the model consistently pick its highest-probability output instead of sampling more varied alternatives, which is exactly what the team needs for a binary classification label that must stay stable across repeated runs on the same input. The maximum token limit adjustment is incorrect because it only controls how long a response can be, and a longer allowed response does nothing to make the underlying word or label choice more deterministic. Adding a content safety filter is incorrect because content safety filters detect harmful content categories like hate speech or violence; they have no role in enforcing consistency of a classification label and would not catch or fix varying urgent or not-urgent outputs. Switching to a larger embedding model is incorrect because embedding models are used to represent text for similarity search or retrieval tasks, not to generate the classification label itself in this chat-style setup, so changing the embedding model would not affect the variability the team is observing. Adjusting temperature is the direct, no-retraining fix for output consistency.