Test Models

Explore how AI models are evaluated on transportation certification exams. Configure the evaluation harness below to see pre-computed results across different datasets, solvers, and scorers.

Test Harness Configuration

Dataset

Solver

System Prompt

You are a transportation professional taking a certification exam. Answer each multiple-choice question by selecting the single best answer. Respond with only the letter of your choice (A, B, C, or D).

Model

GPT 5.2

Scorer

Type

Scoring Script

def exact_match(model_answer: str, correct_answer: str) -> bool:
    """Compare model output to ground truth answer key."""
    return model_answer.strip().upper() == correct_answer.strip().upper()