Skip to main content
Evaluators benchmark your prompt performance on various dimensions, given your test cases. Run tests against your datasets to find what works best for your specific cases. Learn more about each evaluator:

LLM as a Judge

Use LLMs to evaluate your prompts.

JavaScript

Write custom JavaScript code to evaluate prompt responses.

Text Matcher

Validate your prompts with precise text patterns.

Cost

Track the financial impact of your prompts.

Response Length

Test the length of your prompt responses.

Latency

Measure how fast your prompts respond.