OpenDAN-Personal-AI-OS is an open source Python-based Personal AI OS that orchestrates multiple AI modules, with Docker-based deployment and a local knowledge base. Latest activity in 2026-03-28; MIT license; 2055 stars.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
OpenDAN is an open source Personal AI OS that consolidates various AI modules for personal use. It supports AI agents, workflows, environments, marketplaces, and a knowledge base, with a development framework for customizing AI assistants.
How it works
Core concepts include AI Agents driven by LLMs with memory, AI Workflows to organize agents, and an AI Environment that allows access to file systems, IoT devices, and services. It provides a Development Framework to customize agents/workflows and a Knowledge Base integration for personal data.
Getting started
Two installation methods are described: via Docker (recommended) and via source code. Installation steps include pulling the Docker image and running a container with volume mounting for data persistence. The README provides commands:
docker pull paios/aios:latest
docker run -v /your/local/myai/:/root/myai --name aios -it paios/aios:latest
docker start -ai aios
docker start aios
python ./src/srvice/aios_shell/aios_shell.py
There is also a note about installing ffmpeg if needed.
OpenDAN from source (build from repo)
git clone https://github.com/fiatrete/OpenDAN-Personal-AI-OS.git
cd OpenDAN-Personal-AI-OS
virtualenv venv
source ./venv/bin/activate
pip install -r ./src/requirements.txt
Recent releases
- 0.5.1 0.5.1 (MVP) (2024-04-23): The first version of OpenDAN MVP (0.5.1) merged into Master; basic OS framework defined; all-in-one mode.
- 0.0.4 0.05 (2023-06-20): Add
remove_bgfunction; supports removing background from an image via Jarvis. - 0.0.3 0.0.3 (2023-06-16): Using GPT function API
Traction
Stars: 2055; Forks: 217; Open issues: 31
Behind the repo
The project is hosted at fiatrete/OpenDAN-Personal-AI-OS; mentions integration with Docker images and a planned OpenDAN Store in 0.5.2; references CYFS OOD OS as a partial kernel basis in planning.
Caveats
License: MIT. The README notes the project is in very early stages with potential significant changes in the future. OpenDAN 0.5.1 MVP exists; 0.5.2 roadmap includes kernel separation and expanded KB support. ffmpeg may be required if not present on the host.