Ollama is an inference engine for serving models.
You'll need to install Ollama and get it running.
Once you have it running you can use the following to connect it to Bionic.
0.0.0.0
.We need to get Ollama to listen on 0.0.0.0
otherwise services from within k3s
can't connect to it.
Run the following
Get you host with hostname
then curl using that host.
From the models screen you'll need to change the URL from http://llm-api
to http://hostname
. Where host name is the name you get when your run hostname
.