A team selects a model for extracting short warranty clauses. Model A meets accuracy and latency targets at lower cost; model B is larger but offers no measured improvement. Which choice is best supported?

1. Choose model A based on the measured task requirements and continue monitoring.
2. Choose model B solely because more parameters always mean better task results.
3. Alternate models randomly without recording which one handled each request.
4. Reject both because no model can perform extraction.