I run a k8s homelab to surface ollama nodes across multiple machines at home. I have been looking for an aggregation and token tracking proxy. This seems like a good candidate. Happy to contribute docker-build-push pipelines as in #9 and helm charts to run it all in single-server or microservice patterns with split frontend/backend.
I run a k8s homelab to surface ollama nodes across multiple machines at home. I have been looking for an aggregation and token tracking proxy. This seems like a good candidate. Happy to contribute docker-build-push pipelines as in #9 and helm charts to run it all in single-server or microservice patterns with split frontend/backend.