CNCF and Industry Coalition Launch llm-d 1.0: Standardizing Disaggregated Prefill-Decode Serving on Kubernetes
cloud••2 min•cncf.io
Governed under the CNCF and backed by Red Hat, IBM, Google Cloud, CoreWeave, and NVIDIA, llm-d 1.0 establishes a vendor-neutral Kubernetes runtime for distributed AI serving with prefill-decode disaggregation and prefix-cache-aware routing.
Read full article→#CNCF