The open-source language model computer.
Preorder the Light | Get Updates | Documentation
We want to help you build. Apply for 1-on-1 support.
Important
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.
A substantial rewrite to address these concerns and more, including the addition of RealtimeTTS and RealtimeSTT, is occurring here.
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 staying open, modular, and free.
git clone https://github.com/OpenInterpreter/01 # Clone the repository
cd 01/software # CD into the source directorybrew 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)The RealtimeTTS and RealtimeSTT libraries in the incoming 01-rewrite are thanks to the state-of-the-art voice interface work of Kolja Beigel. Please star those repos and consider contributing to / utilizing those projects!
- The 01 Light is an ESP32-based voice interface. Build instructions are here. A list of what to buy here.
- It works in tandem with the 01 Server (setup guide below) running on your home computer.
- Mac OSX and Ubuntu are supported by running
poetry run 01(Windows is supported experimentally). This uses your spacebar to simulate the 01 Light. - (coming soon) The 01 Heavy is a standalone 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 →
The 01 exposes a speech-to-speech websocket at localhost:10001.
If you stream raw audio bytes to / in Streaming LMC format, you will receive its response in the same format.
Inspired in part by Andrej Karpathy's LLM OS, we run a code-interpreting language model, and call it when certain events occur at your computer's kernel.
The 01 wraps this in a voice interface:
To communicate with different components of this system, we introduce LMC Messages format, which extends OpenAI’s messages format to include a "computer" role:
LMC.mov
Dynamic System Messages enable you to execute code inside the LLM's system message, moments before it appears to the AI.
# Edit the following settings in i.py
interpreter.system_message = r" The time is {{time.time()}}. " # Anything in double brackets will be executed as Python
interpreter.chat("What time is it?") # It will know, without making a tool/API callTo run the server on your Desktop and connect it to your 01 Light, run the following commands:
brew install ngrok/ngrok/ngrok
ngrok authtoken ... # Use your ngrok authtoken
poetry run 01 --server --exposeThe final command will print a server URL. You can enter this into your 01 Light's captive WiFi portal to connect to your 01 Server.
poetry run 01 --local
If you want to run local speech-to-text using Whisper, you must install Rust. Follow the instructions given here.
To customize the behavior of the system, edit the system message, model, skills library path, etc. in i.py. This file sets up an interpreter, and is powered by Open Interpreter.
sudo apt-get install portaudio19-dev ffmpeg cmakePlease see our contributing guidelines for more details on how to get involved.
Visit our roadmap to see the future of the 01.
The story of devices that came before the 01.
Things we want to steal great ideas from.
○