Introduction

Empirical is the fastest way to test different LLMs and model configurations, across all the scenarios that matter for your application.

Quick start →

​

With Empirical, you can

• Run your test datasets locally against off-the-shelf or custom models
• Compare model outputs on a web UI, and test changes quickly
• Score your outputs with scoring functions
• Run tests on CI/CD

​

Walk through

Watch a demo video showing Empirical run a text-to-SQL example with TypeScript config.

​

Open source

Empirical is open source on GitHub. Star the repo, file issues or pull requests to contribute to the project.