Today, OpenAI released its latest LLM model, GPT-4o. People are sharing crazy applications built on top of this groundbreaking model. By using GPT-4o, we see improvements in data retrieval, which could benefit users with faster response and be more cost-efficient (Cost less than GPT -4 Turbo).

Within 12 hours, our team at Wren AI released our latest version, to support the GPT-4o LLM model for querying your own database. In this post, we will guide you through setting up using gpt-4o to query your own PostgreSQL database!

Wren AI is an open-source text-to-SQL solution for data teams to get results and insights faster by asking business questions without writing SQL.

Let’s get started!

Setup your PostgreSQL database

Start your PostgreSQL server, and enter your PostgreSQL database using the psql client.

Create a database in PostgreSQL here, we called ecom-wrenai

CREATE DATABASE ecom_wrenai;

Connect to the created database ecom_wrenai using \c <database_name>

lijungchi=# \c ecom_wrenaipsql (13.1, server 13.0)You are now connected to database "ecom_wrenai" as user "lijungchi".

ecom_wrenai=#

Now, you want to create tables of this dataset and import the data. Please check the SQL script here in the gist.

You can either copy the script to your psqlor Import SQL via psql

ecom_wrenai=# \i <path to your file>/import.sql

If you successfully import the data, you will see the information below.

Installing Wren AI

Next, let’s start installing Wren AI; before we start, you need to install Docker.

1. Install Docker Desktop on your local computer.​

Please ensure the version of Docker Desktop is at least >= 4.17.

2. Prepare an OpenAI API key​

Please ensure that your Open API key has Full Permission(All).

Visit the OpenAI developer platform.

Generate a new API key for Wren AI with full permission

3. Install Wren AI Launcher

If you are on Mac(using Windows or Linux check here) enter the below command to install the latest Wren AI Launcher.

curl -L https://github.com/Canner/WrenAI/releases/latest/download/wren-launcher-darwin.tar.gz | tar -xz && ./wren-launcher-darwin

The launcher will then ask for your OpenAI API key as below, paste your key into the command and hit enter.

Now you can select gpt-4o , gpt-4-turbo , gpt-3.5-turbo of OpenAI’s generation model in Wren AI.

Now you’ll see we are running docker-compose on your computer; after the installation, the tool will automatically open up your browser to access Wren AI.

Connect your Data Source

While the terminal is successfully installed, it will launch the browser

Select PostgreSQL database connector

If you use the Mac local PostgreSQL database, enter docker.for.mac.localhost to access the database.

Next step, define relationships to the semantic models; this could help LLMs to understand more about your semantic data structure.

Now, you are all set!

You can view all your data models in the Wren AI UI.

Playing around with Wren AI

Wren AI UI semantic modeling

With Wren AI UI, you can model your data models in a semantic context, such as adding descriptions, relationships, calculations, etc. The context could help LLMs learn your business terminologies and KPI definitions and reduce hallucinations when joining multiple tables; LLMs know the data structure hierarchy by learning through relationships such as whether it is a many-to-one , one-to-many , or many-to-many relationships between tables.

Define your business KPIs and formulas via calculations in Wren AI.

Adding semantic relationships between tables.

Ask business questions and Follow-up insights with Wren AI

Now you can switch to the Home page in Wren AI UI, by clicking New Thread you can start asking any of your business questions, Wren AI search for relevant semantics through the vector store and provides recommendations.

Wren AI will show up to 3 options for your business questions; you can select the most relevant question to generate the final result.

As you select an option, Wren AI will display a step-by-step breakdown of the summary, including SQL and data, allowing you to verify if the result is what you were looking for.

You can ask follow-up questions to gain deeper insights into the result.

Let’s about it! Hope you are having fun with gpt-4o!

‍