Red Hat AI Inference on CKS
CoreWeave Blog, Tuesday, May 12th, 2026
CoreWeave and Red Hat announce a deployment blueprint for running Red Hat AI Inference on CoreWeave Kubernetes Service for hybrid enterprise inference workloads.
CoreWeave and Red Hat have announced a new deployment blueprint that enables enterprise teams to run Red Hat AI Inference on CoreWeave Kubernetes Service (CKS) in hybrid on-premises and cloud environments. The solution provides a tested, documented reference architecture that allows teams to use the same open-source inference stack across environments while maintaining Kubernetes-native control and operational consistency.
Red Hat AI Inference includes model serving gateways, distributed LLM serving through the llm-d project, support for multiple inference servers and accelerators, and various optimization techniques. CKS is purpose-built for AI with deep observability, automated node lifecycle management, and access to NVIDIA's latest GPU generations and InfiniBand networking, making it an ideal foundation for production inference workloads.