-
Run the code-llama-2 LLM in Openshift
-
Requires a worker node with NVIDIA GPU support
-
Create the Openshift application
oc new-app python~https://github.com/bkoz/code-llama-server
- Expose the Gradio service as a route.
oc create route edge code-llama-server --service=code-llama-server --insecure-policy='Redirect'