EngiBench: A Benchmark for Evaluating Large Language Models on Engineering Problem Solving

Understand

Large language models (LLMs) have shown strong performance on mathematical reasoning under well-defined conditions.

• However, real-world engineering problems involve uncertainty, context, and open-ended settings that extend beyond symbolic computation.
• Existing benchmarks largely focus on well-defined or abstract reasoning and therefore fail to capture these complexities.
• We introduce EngiBench, a hierarchical benchmark designed to evaluate LLMs on solving engineering problems.