benchmarking large language models (LLMs) with a focus on their mathematical capabilities - View it on GitHub