Using Ollama UBI as internal k8s service

https://github.com/williamcaban/ollama-ubi

with Ollama UBI pod, pull the required model

This command will download the mistral model for ollama to use

$ ollama pull mistral

This command will run mistral at http://localhost:11434/v1

ollama run mistral

Build chat image

$ podman build -f Containerfile -t quay.io/rhn_support_arolivei/chat

deploy on k8s/microshift/openshift

oc apply -f manifests/02-chat-serve-deployment.yaml
oc apply -f manifests/03-chat-svc.yaml
oc apply -f manifests/04-chat-route.yaml