More often than not we pre-process data using an LLM before inserting them into a column/row on the database, for example translating text and storing the different languages or running OCR on a document and storing the structured output or using Classification to label user messages/documents before storing it.

Many of this use cases could happen on the database rather than a dedicated pipeline. With the motivation to allow Postgres to do more by extending it. We built Postgres-LLM which is a trigger than can conditionally be set on specific columns and executes a well defined tasks on insert/update that runs an LLM and stores the output on another column or updates the initial column.

This project supports any LLM that is OpenAI Chat API compatible. I was motivated to create this project to showcase our recently launched LLM, Interfaze.ai as it was trained to work on developer tasks like OCR, translation, classification and more. However, you can use any LLM of your choice by replacing the URL and API key on setup.