Announcing llm-d 1.0: Open distributed inference for agents on Google Cloud TPUs and GPUs

Google, in collaboration with Red Hat, IBM, and open-source contributors, has launched the 1.0 release of llm-d under the Cloud Native Computing Foundation Sandbox. The framework provides a Kubernetes-native foundation designed to orchestrate and scale distributed model inference for agentic systems across both GPUs and TPUs. The system unifies model-serving components including vLLM and SGLang with enterprise container management infrastructure.