Local Agent platform with generative AI models and tools to make AI helpful for everyone.
Before installing Argo, make sure your machine meets the following minimum system requirements:
- CPU >= 2 Core
- RAM >= 16 GB
- Disk >= 50 GB
- GPU >= 8G(Mac M1 above)
- Extra Software requirements with Docker
- Docker >= 24.0.0
Warning
To enable CUDA in Docker, you must install the Nvidia CUDA container toolkit on your Linux/WSL system.
Download, Click and Install.
- Macos silicon(M1 and above):argo-1.0-osx-installer.dmg
- Windows 64bit(win 10 and above):argo-1.0-windows-x64-installer.exe
Quick start with Docker 🐳
- If you are using Linux, Ollama included in the image by default.
- If you are using MacOS (Monterey or later), Ollama deployed on the host machine by default.
- If you are using Windows, please manually install Docker and WSL environment, then follow Linux instructions.
- We'll use brew to install docker and ollama, if something wrong, you can install docker and ollama yourself.
Tip
Recommend Ollama model: glm4
for chat, shaw/dmeta-embedding-zh
for Knowledge of chinese.
# Usage: {run [-n name] [-p port] | stop [-n name] | update}
# default name: argo
# default port: 38888
# Download image, create a container and start
sh argo_run_docker.sh run
# Stop the container (data will be retained)
sh argo_run_docker.sh stop
# Update the image to the latest version (the original image will be deleted)
sh argo_run_docker.sh update
Free to join us and talk: https://discord.gg/79AD9RQnHF
Let's make Argo even more amazing together! 💪