Key Features • Architecture • Installation Guide • Community • Acknowledgements • License
- Multi-lingual turnkey solution for GenAI use, development and deployment on Windows, Linux and MacOS
- Concurrent multi-chat, quoting, full prompt-list import/export/share, and more for users
- Supporting multimodal models, popular RAG/agent tools, traditional applications, and local bot store
- Flexible orchestration of prompts x RAGs x multi-modal models x tools x bots x hardware/GPUs
- Heterogeneous support from raspberry Pi, laptops, PCs, edge servers, and virtual hosts to cloud
- Open-sourced, allowing developers to contribute and customize the system according to their needs
Warning: This a preliminary draft and may be subject to further changes.
Download the script or the executable file, run it, and follow its steps to have your own Kuwa!
-
Windows
Download and run the pre-built Windows executable from Kuwa's latest releases
-
Linux/Docker
Download and run sudo build.sh , or invoke the following command to automatically install Docker, CUDA, and Kuwa. You may need to reboot after installing CUDA. Before finishing installation, you will be asked to set your administration passwords for your Kuwa and database. After installation, it will invoke run.sh to start the system and you can log in with admin@localhost. Enjoy!
curl -fsSL https://raw.githubusercontent.com/kuwaai/genai-os/main/docker/build.sh | sudo sh
You can build your own customized Kuwa by following the step-by-step documents.
With executors, Kuwa can orchestrate diverse multimodal models, remote services, applications, databases, bots, etc. You can check Executor's README for further customization and configuration.
You can download the latest Kuwa GenAI OS version that supports Windows and Linux.
Discord - Kuwa AI Discord community server
Facebook - Kuwa AI Community
Facebook - Kuwa AI Taiwan community
Google Group - kuwa-dev
Facebook - Kuwa AI
Google Group - kuwa-announce
If you are interested in this project, you can help us develop it together and improve this open-source project. Please do not hesitate to contact us anytime if you are willing to help!
The following packages and applications are used in this project:
We want to acknowledge NSTC's TAIDE project and the Taiwan AI Academy for their assistance in the early development of this project.