Find the best model and prompt for your application

Empirical is the fastest way to try different LLMs and prompt combinations, across all the scenarios that matter.

Define your input scenarios, select a model, and hit run.

Explore comparisons

Text-to-SQL

Compare LLMs on converting natural language questions to SQL queries on the Spider dataset.

Select two models to compare

gpt-3.5-turbo

gpt-4

gpt-4-1106-preview

gpt-4-turbo-preview

gemini-pro

gemma-7b-it

gemma-2b-it

mistral-large-latest

mistral-tiny

mistral-small

llama-v2-7b-chat

claude-3-sonnet

claude-2.1

claude-3-haiku

From our blog

Feb 3, 2024

The AI Developer's Handbook: Practical guide to evaluate AI-generated SQL

Jan 30, 2024

The AI Developer's Handbook: Comparing Models

Jan 18, 2024

Introducing Empirical: a playground for LLM comparisons

Dec 30, 2023

Building for the AI engineer

Subscribe to updates

Stay updated on our progress with demos sent to your inbox.

About us

Empirical is founded by Arjun Attam and Saikat Mitra. We are a small team of dev tool builders who believe that LLMs will transform app development, and consequently, the lives of end-users. Our goal is to empower millions of app developers with our work, enabling them to lead this transformation.