Maxim provides a collection of pre-built evaluators that you can use immediately for your AI evaluation needs. These include high-quality evaluators from Maxim and popular open-source libraries like RAGAS. Install them directly from the Evaluator Store.

The Evaluator Store contains three types of evaluators:

  • AI-based - Uses LLMs as judges with well-curated prompts (Example: Clarity evaluator)
  • Programmatic - Uses JavaScript or Python code to evaluate quality (Example: isValidJSON)
  • Statistical - Traditional ML metrics for text comparison (Example: BLEU, ROUGE)

1

​
Install evaluators

Search or filter evaluators based on your requirements. Add them to your workspace by clicking the Add to workspace button.

2

​
Use installed evaluators

Use your installed evaluators while triggering test runs on any entity.