unbody is a free, open source machine learning infrastructure project written in TypeScript and released under Apache-2.0. It has 524 GitHub stars, 47 forks and 3 open issues, and was last pushed 5 months ago. On this registry it ranks #81 of 86 tracked projects in Machine Learning Infrastructure, with 5 head-to-head comparisons available.

Unbody (archived)

This repository is archived and no longer actively maintained.

Unbody started as an open-source project to build the Supabase of the AI era. That vision has evolved.

We now build under Unbody Labs — shipping focused tools and products for the AI world. Our most active and direct continuation of Unbody's mission is Adapt.

Adapt

Adapt is a lightweight (

Original setup instructions

Prerequisites

  • Node.js LTS (20 or 22)
  • Docker and Docker Compose
  • yarn (npm won't install dependencies correctly)
  • OpenAI API key

Installation

git clone https://github.com/unbody-io/unbody
cd unbody
yarn

Environment Setup

cp .env.example .env.local
vim .env.local

Running the Application

docker compose up -d
yarn start

Create Your First Project

git clone https://github.com/unbody-io/examples.git
cd examples

Add a Data Source

yarn unbody-cli source add

Monitor Indexing Progress

Head over to the Temporal dashboard at http://localhost:8233/ to see your files being indexed.

  • The dashboard needs to be manually refreshed to get the latest state
  • Once indexing is complete, follow the README from the examples repository

Frequently asked questions

Is unbody free to use?

unbody is open source under the Apache-2.0 licence. There is no licence fee and no seat count — you can self-host it or, where the project offers one, pay a vendor for a managed version instead.

What does unbody do?

The Supabase of AI era. A modular, open-source backend for building AI-native software — designed for knowledge, not static data.

What is unbody written in?

unbody is primarily written in TypeScript. Its source is publicly available at https://github.com/unbody-io/unbody, and it has 524 GitHub stars.