WebCoderBench: Benchmarking Web Application Generation with Comprehensive and Interpretable Evaluation Metrics

Researchers introduced WebCoderBench, the first comprehensive benchmark for evaluating web application generation by large language models, featuring 1,572 real-world user requirements and 24 evaluation metrics. The benchmark tests 12 representative LLMs and shows no single model dominates across all metrics, providing opportunities for targeted improvements.

• →WebCoderBench is the first real-world benchmark specifically designed for evaluating LLM web application generation capabilities.
• →The benchmark includes 1,572 authentic user requirements covering diverse modalities and expression styles.
• →It provides 24 fine-grained evaluation metrics across 9 perspectives using both rule-based and LLM-as-a-judge paradigms.
• →Testing of 12 representative LLMs revealed no dominant model across all evaluation criteria.
• →The benchmark offers LLM developers clear opportunities for targeted model optimization in web development tasks.