This report evaluates 41 open-source large language models using 19 benchmark tests, showcasing their performance across various tasks.