This repository contains benchmark datasets (images and text), prompts, ground truths, and evaluation scripts for assessing the performance of large language models (LLMs) on humanities-related tasks. The suite is designed as a resource for researchers and practitioners interested in systematically evaluating how well various LLMs perform on digital humanities (DH) tasks involving visual and text-like materials. For detailed test results and model comparisons, visit our results dashboard at https://rise-services.rise.unibas.ch/benchmarks/.

If you use this software, please cite it using the metadata from this file.