exo: Run your own AI cluster at home with everyday devices. Maintained by exo labs.

Discord | Telegram | X

---

Forget expensive NVIDIA GPUs, unify your existing devices into one powerful GPU: iPhone, iPad, Android, Mac, Linux, pretty much any device!

Get Involved

exo is experimental software. Expect bugs early on. Create issues so they can be fixed. The exo labs team will strive to resolve issues quickly.

We also welcome contributions from the community. We have a list of bounties in this sheet.

Features

Wide Model Support

exo supports LLaMA and other popular models.

Dynamic Model Partitioning

exo optimally splits up models based on the current network topology and device resources available. This enables you to run larger models than you would be able to on any single device.

Automatic Device Discovery

exo will automatically discover other devices using the best method available. Zero manual configuration.

ChatGPT-compatible API

exo provides a ChatGPT-compatible API for running models. It's a one-line change in your application to run models on your own hardware using exo.

Installation

The current recommended way to install exo is from source.

From source

git clone https://github.com/exo-explore/exo.git
cd exo
pip install -r requirements.txt

Documentation

Example Usage on Multiple Devices

Device 1:

python3 main.py

Device 2:

python3 main.py

That's it! No configuration required - exo will automatically discover the other device(s).

A ChatGPT-like web interface will be available on each device on port 8000 http://localhost:8000.

An API endpoint will be available on port 8001. Example usage:

curl -X POST http://localhost:8001/api/v1/chat -H "Content-Type: application/json" -d '{"messages": [{"role": "user", "content": "What is the meaning of life?"}]}'

Inference Engines

exo supports the following inference engines:

• MLX
• tinygrad
• 🚧 llama.cpp

Networking Modules

• ✅ GRPC
• 🚧 Radio
• 🚧 Bluetooth