Red Hat releases AI 3.4 for AI agents on hybrid cloud
Red Hat tells that AI agents will cause more use of inference, the operation of models to give answers.
At Red Hat Summit in May 2026, Red Hat showed how it helps companies move AI into production at scale. It released Red Hat AI 3.4 for AI agents on hybrid cloud, a mix of company systems and public cloud. Red Hat also released two managed services and Ansible Automation Platform 2.7, and it gave no test results.
What is new
- Red Hat AI Inference puts real time inference in the workflows of a company.
- Red Hat OpenShift Virtualization Service on IBM Cloud operates virtual machines at scale.
- Ansible 2.7 connects AI results to controlled steps on servers and networks.
Red Hat tells that a step by a human or an AI agent uses the same access controls.
The numbers
Analysts of theCUBE Research tell that 50.7% of organizations use public AI tools. Only 20.2% have AI used in all of the company on a governed framework.
They also tell that AI cloud spending can come near $2.5 trillion by 2026.
What is not known
The source gives no prices and no test results. AMD and Intel tell that the right mix of hardware can lower cost. They are partners of Red Hat.
Insight: Red Hat works with three hardware suppliers
Red Hat works with three hardware suppliers at the same time. On 9 November, Nvidia showed a stack that selects models by cost. This reads as a play to be the neutral layer between hardware suppliers and agents. Buyers must find the cost to change hardware.
Sources
Posted
