Evaluating Open Source Large Language Models
By Shreya Pandey and Rachel Hines, guided by Daniele Di Sarli
Large Language Models (LLMs) have become a popular tool for semantic text processing. Rachel Hines and Shreya Pandey present a comprehensive analysis of the performance of several LLMs on different semantic tasks and categories, such as reasoning, summarization, bias and text formatting. They also extend the analysis with information on speed and resource usage for each of the LLMs that are analyzed. Each LLM under study shows different peculiarities, making the choice of the best LLM high dependent on the use case. They provide guidance on choosing LLMs.
