LLM as a Judge

The standard for evaluating text is human labeling. However, high-quality LLM outputs are becoming cheaper and faster to produce, and human evaluation cannot scale. In this context, evaluating the performance of LLM applications is best tackled by using a LLM. The Phoenix LLM Evals library is designed for simple, fast, and accurate LLM-based evaluations.

Learn more

Test LLM evaluation in playground

Run evaluations in the UI

Run evaluations with code

Learn about LLM evaluation concepts

To get started, check out the Quickstart guide for evaluation. You can use Arize Evaluation templates or build your own. Read our best practices to learn how to run robust task-based evaluations on your LLM applications. Finally, check out our comprehensive guide to building and benchmarking LLM evals.