Skip to content

Latest commit

 

History

History
224 lines (132 loc) · 6.67 KB

README.md

File metadata and controls

224 lines (132 loc) · 6.67 KB

Official pre-release repository for The 01 Project.

1 day remaining until launch



Discord

The open-source language model computer.


poster

We want to help you build. Apply for 1-on-1 support.



⚠️ WARNING: This experimental project is under rapid development and lacks basic safeguards. Until a stable 1.0 release, ONLY run this repository on devices without sensitive information or access to paid services. ⚠️



The 01 Project is building an open-source ecosystem for AI devices.

Our flagship operating system can power conversational devices like the Rabbit R1, Humane Pin, or Star Trek computer.

We intend to become the GNU/Linux of this space by committing to staying open, modular, and free.


Software

git clone https://github.com/OpenInterpreter/01 # Clone the repository
cd software/source # CD into the source directory

Not working? Read our setup guide.

brew install portaudio ffmpeg cmake # Install Mac OSX dependencies
poetry install # Install Python dependencies
export OPENAI_API_KEY=sk... # OR run `poetry run 01 --local` to run everything locally
poetry run 01 # Runs the 01 Light simulator (hold your spacebar, speak, release)

Hardware

  • The 01 Light is an ESP32-based voice interface. Build instructions are here. It works in tandem with the 01 Server running on your home computer.
  • Mac OSX and Ubuntu are supported by running poetry run 01. This uses your spacebar to simulate the 01 Light.
  • The 01 Heavy is a device that runs everything locally.

We need your help supporting & building more hardware. The 01 should be able to run on any device with input (microphone, keyboard, etc.), output (speakers, screens, motors, etc.), and an internet connection (or sufficient compute to run everything locally). Contribution Guide →


How does it work?

The 01 exposes a speech-to-speech websocket at localhost:10001.

If you stream raw audio bytes to / in LMC format, you will recieve its response in the same format.

Inspired in part by Andrej Karpathy's LLM OS, we point a code-interpreting language model at your computer's kernel, forming a language model computer (LMC).


LMC

This architecture fuses classical computers (powerful, deterministic) with language models (flexible, non-deterministic). We believe it inherits the best of both— unifying the unreasonable power of classical computers with the human-like usability of language models.




Protocols

LMC Messages

Dynamic System Messages

The 01OS

The 01OS can be housed in many different bodies. We highly encourage PRs that add to this list:

The 01 Light is an ESP32-based voice interface that controls your home computer over the internet. It's used in combination with the 01 Server.

The 01 Heavy is a device that runs everything locally.

Software

Install dependencies

# MacOS
brew install portaudio ffmpeg cmake

# Ubuntu
sudo apt-get install portaudio19-dev ffmpeg cmake

If you want to run local speech-to-text using Whisper, install Rust. Follow the instructions given here.

Install and run the 01 CLI

pip install 01OS
01 --server # Start a server for a hardware device to listen to.

Client Setup

For ESP32 boards

Please visit our ESP32 setup documentation.

For Mac, Windows, and Ubuntu machines

01 # Start a server and a client.

01 --server --expose # Start and expose a server via Ngrok. This will print a `server_url` for clients to connect to.

01 --client --server_url your-server.com # Start only a client.

Swap out service providers

The 01 is model agnostic to speech-to-text, text-to-speech, and language model providers.

Select your provider by running:

01 --tts-service openai
01 --llm-service openai
01 --stt-service openai

View all providers ↗, or join the 01 team by adding a service provider. ↗

Run the 01 locally

Some service providers don't require an internet connection.

The following command will attempt to download and use the best providers for your hardware:

01 --local

How Does it Work?

The 01 equips a language model (wrapped in a voice interface) with an exec() function, which allows it to write and run code to control computers.

We only stream speech to/from the end user's device.

Contributing

Please see our contributing guidelines for more details on how to get involved.

Setup for development

# Clone the repo
git clone https://github.com/KillianLucas/01.git

# Go to the 01OS directory
cd 01OS

# Install python dependencies
poetry install

# Run it
poetry run 01

Roadmap

Visit our roadmap to see the future of the 01.


Background

The story of devices that came before the 01.

Things we want to steal great ideas from.


Direction

What we're going to do.

What the 01 will be able to do.