n8n - Self-hosted AI Starter Kit
-
The Self-hosted AI Starter Kit is an open-source template that quickly sets up a local AI environment. Curated by n8n, it provides essential tools for creating secure, self-hosted AI workflows.
https://github.com/n8n-io/self-hosted-ai-starter-kit
I believe this could be very interesting to support. n8n is already supported, but this tool seems like a completely different tool, with more pieces.
Already supports Docker Compose, so it seems it could be possible to add it as a new Cloudron App.
I'd typically want to spawn a new server and start testing things without bothering with all the installation details.
Also, providing some tips about optimizing its performances (using "Nvidia GPU") would be very helpful when selecting the server to host this app. I believe Cloudron might have a few partnerships with Cloud providers to help us pick a good choice for improved performances out of the box.
-
While this kit is more geared towards folks with a local GPU, the idea behind is not bad at all.
The guy behind bolt.diy has a great video on how to use the new features of n8n which include AI Agent nodes which you can connect to any external API.
Here's the video:
I also think this is a nice way to train the Agent to analyze Cloudron repos of packaged Apps to learn how to package new ones with less effort.
Give it a go!