MingZhong/

crosseval_scores

	📜 CrossEval: Benchmarking LLM Cross Capabilities
















	


	Release of Model Responses and Evaluations

In addition to the CrossEval benchmark, we release the responses from all 17 models in this repository, along with the ratings and explanations provided by GPT-4o as the evaluator. The included model families are:

GPT Claude Gemini Reka

Each instance in the dataset contains the following fields:

prompt_id:… See the full description on the dataset page: https://huggingface.co/datasets/MingZhong/crosseval_scores.