D
DeepEval is an open-source LLM evaluation framework for testing and benchmarking AI applications. It helps developers run pytest-native evaluations, score outputs and agent traces, and iterate on systems across text, image, audio, and voice workflows.
免費及付費方案