KServe Provides NIMble Cloud & Data Centre Inference

NVIDIA NIM, a set of microservices for generative AI inference, will connect with KServe, an open-source programme that automates cloud computing AI model deployment

KServe was part of Kubeflow, an open-source machine learning toolkit based on Kubernetes, a containerisation solution for large distributed applications

KServe makes it simple for users to switch between models to see which one best meets their requirements

KServe feature known as “canary rollouts” automates the process of meticulously validating and progressively releasing an updated model into production when one is available

Due to its interface with KServe, NIM will be available on various corporate platforms, including Red Hat's OpenShift AI, Canonical's Charmed KubeFlow and Kubernetes, Nutanix GPT-in-a-Box 2.0, and others

Red Hat and NVIDIA are making open source AI deployment easier for enterprises

NVIDIA NIM inference microservices will enable consistent, scalable, secure, high-performance generative AI applications from the cloud to the edge