Things built to be run
klah.io builds and publishes tooling for on-prem Kubernetes. What lands here is the part of that work that holds up outside the cluster it was written for: charts released so anyone can install them, read the defaults, and see why each one is what it is.
Each project is Apache-2.0 on GitHub, versioned, and installable with a single command. Nothing here is a demo. If one of them breaks on your cluster, that is a bug worth an issue.
ai-stack-lite
ai-stack-lite
A self-hosted AI stack for an on-prem Kubernetes cluster, in one chart: a local LLM engine, a chat frontend, and an image-generation workflow runner. Each piece is its own subchart and can be switched off alone, and a shared library chart keeps naming, labels, scheduling, persistence and GPU handling under one set of rules instead of three.
- Ollama
- LLM inference engine, holds the model weightscluster-internal
- Open WebUI
- chat frontendIngress or HTTPRoute
- ComfyUI
- image generation workflowsport-forward
# install from the published repository
helm repo add ai-stack-lite https://klahio.github.io/ai-stack-lite
helm repo update
helm install ai ai-stack-lite/ai-stack-lite -n ai-stack --create-namespace
Profiles ship for the two common shapes: values-single-gpu.yaml for one
consumer GPU shared by Ollama and ComfyUI, and values-cpu.yaml for a
cluster with no GPU at all.
The chart does not install GPU support, on purpose. A device plugin or the GPU
Operator is node-level infrastructure with its own lifecycle, and bundling it
would mean a helm uninstall of this stack takes the cluster's GPU
support with it. The chart checks the prerequisites at install time and fails
with instructions instead.
Contact
Bugs and feature requests belong in the issue tracker of the repository they concern — that keeps the answer where the next person will look for it. For anything else, write to [email protected].