Skip to content
FindOpenSource

DeepEval

Open-source testing framework for evaluating LLM outputs, treating LLM application quality checks like conventional unit tests.

Overview

DeepEval provides metrics and test primitives specifically for LLM outputs - answer relevancy, faithfulness to source documents, hallucination detection, and more - so a team can write automated tests for an LLM application's quality the same way they'd write unit tests for regular code, and run them in CI. It integrates with common testing frameworks like Pytest.

DeepEval is a fit for teams that want systematic, automated quality checks on LLM or RAG application outputs rather than manual spot-checking, complementing Promptfoo (also in this catalog) - both address LLM evaluation, with DeepEval leaning toward a pytest-native unit-testing workflow.

Categories
AI & Machine Learning
Keywords
llm-evaluationtestingrag-evaluation
Languages
Python
License
Apache-2.0

Spotted an error? Suggest an edit on GitHub.