ome

Open Model Engine (OME) — Kubernetes operator for LLM serving, GPU scheduling, and model lifecycle management. Works with SGLang, vLLM, TensorRT-LLM, and Triton. Maintained by ome-projects on GitHub, where it has earned 480 stars from the community.

It's actively developed around deepseek, k8s, kimi-k2, and is a solid reference for anyone building with these tools.