Business news |
By Brian Tristam Williams
IBM has announced two new managed cloud services built around Red Hat technology, adding Red Hat AI Inference on IBM Cloud and Red Hat OpenShift Virtualization Service on IBM Cloud to its enterprise cloud portfolio.
The company said the services are intended to help customers move AI workloads from pilot projects into production, while also giving enterprises a managed route for running virtual machine workloads on OpenShift. IBM announced two managed services on May 12, 2026, positioning them as part of its wider hybrid cloud and AI strategy.
The move follows IBM’s long-running effort to tie its enterprise AI plans to Red Hat’s open hybrid cloud stack. As eeNews Europe reported when IBM agreed to acquire Confluent, the company has been expanding around enterprise data, integration and AI infrastructure rather than treating AI as a standalone software layer.
Red Hat AI Inference on IBM Cloud is a fully managed inference service designed to run production AI models without customers having to manage GPUs, runtime infrastructure or the AI platform underneath. IBM says the service uses Red Hat AI’s inference engine, with vLLM, to support low-latency and high-throughput model serving for real-time and agentic AI workloads.
The service includes OpenAI-compatible APIs, IBM Cloud IAM integration, audit logging, privacy controls and service-level reliability features. The model catalogue listed by IBM includes Granite 4.0 H Small, Mistral-Small-3.2-24B-Instruct, Llama 3.3 70B Instruct, GPT-OSS-120B and Nemotron-3-Nano-30B-FP8, with further open and custom models planned.
Red Hat describes its inference stack as an integrated platform for running models across hybrid cloud environments, using vLLM and llm-d to improve response times, hardware use and token economics. The managed IBM Cloud version is aimed at organisations that want model-as-a-service deployment without building and operating the inference layer themselves.
The second service, Red Hat OpenShift Virtualization Service on IBM Cloud, gives enterprises a managed option for migrating and operating VM-based workloads on Red Hat OpenShift. It runs on IBM Cloud VPC Bare Metal and uses Kubernetes-based infrastructure, automated lifecycle management and migration tooling, including the Migration Toolkit for Virtualization.
IBM said the service is intended for companies reassessing their virtualisation platforms and looking for a more predictable operating model. It also gives customers a route to run VMs and containers on the same OpenShift foundation, rather than maintaining separate platforms for legacy and cloud-native workloads.
IBM said Jason McGee, CTO of IBM Cloud, described Red Hat AI Inference on IBM Cloud as a managed platform “built for real workloads, not just experiments.” Red Hat chief product officer Ashesh Badani said the services are part of Red Hat and IBM’s work to support an open, consistent hybrid cloud platform for enterprise AI.
Red Hat AI Inference on IBM Cloud is scheduled for general availability on May 22, 2026. Red Hat OpenShift Virtualization Service on IBM Cloud is currently in limited availability, with general availability expected in June 2026. IBM noted that features and timelines remain subject to change and may not be available in all countries.
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。