Details
Confident AI offers DeepEval, an open-source LLM evaluation framework, and a managed platform for systematic LLM testing. DeepEval provides Python-based evaluation similar to pytest with metrics for hallucination, relevance, faithfulness, and many other dimensions.