Projects by llm-d

Achieve state of the art inference performance with modern accelerators on Kubernetes