Alpha Excel Benchmark
This work provides a standardized framework for assessing LLM capabilities in practical business applications, bridging the gap between academic benchmarks and real-world use for the 1.5 billion daily Excel users, though it is incremental as it adapts existing challenges into a new format.
The study tackled the problem of evaluating Large Language Models (LLMs) in realistic business tasks by creating a benchmark from 113 Financial Modeling World Cup Excel challenges, finding significant performance variations across categories with models excelling in pattern recognition but struggling with complex numerical reasoning.
This study presents a novel benchmark for evaluating Large Language Models (LLMs) using challenges derived from the Financial Modeling World Cup (FMWC) Excel competitions. We introduce a methodology for converting 113 existing FMWC challenges into programmatically evaluable JSON formats and use this dataset to compare the performance of several leading LLMs. Our findings demonstrate significant variations in performance across different challenge categories, with models showing specific strengths in pattern recognition tasks but struggling with complex numerical reasoning. The benchmark provides a standardized framework for assessing LLM capabilities in realistic business-oriented tasks rather than abstract academic problems. This research contributes to the growing field of AI benchmarking by establishing proficiency among the 1.5 billion people who daily use Microsoft Excel as a meaningful evaluation metric that bridges the gap between academic AI benchmarks and practical business applications.