Extracting structured information from documents with LLM

This repository contains code from the blog post about using Batch API to reduce your LLM costs by 50%.

OpenAI API key

In order to run this project, you need to create an API key in OpenAI:

• Run cp .env.sample .env
• Create an account or log in with you existing one
• Where do I find my API key?
• Create a new API key
• Copy the key and paste it in the .env file

Setup

This project is meant to be run with Docker and Devcontainers. The settings are located inside .devcontainer/devcontainer.json.

To set up a devcontainer with your IDE, follow these instructions:

• For VSCode
• For IntelliJ

Once the devcontainer is created, run a terminal inside it and execute:

• uv sync

Setup without Devcontainer

Follow these steps:

• Install Python v3.12
• Install uv
• Run uv sync

Running the project

Run uv run create_batch.py