Red Hat Released Red Hat AI 3.5
The software update introduces advanced safety, observability, and agentic development tools for enterprise workloads.
Updated on Sept. 29, 2026 in Artificial Intelligence
Live Poll
Do you trust that businesses have the necessary controls to deploy AI safely at scale?
Red Hat has launched version 3.5 of its Red Hat AI platform. The update focuses on operational rigor by adding new features for model verification, multi-tenancy support, and performance monitoring.
Why it matters
Organizations often struggle to move artificial intelligence projects from pilot phases to production. This release provides the necessary observability and safety tools to help companies scale these workloads reliably.
Red Hat AI 3.5 introduces EvalHub for regulatory compliance and inference-time scaling that adjusts compute based on query complexity. The platform now supports multi-tenancy on shared GPU infrastructure via Red Hat OpenShift hosted control planes.
The players
Red Hat
Red Hat is a prominent provider of open-source software solutions, including the OpenShift container platform used for managing enterprise applications.
The details
The update adds observability dashboards that track critical metrics like GPU utilization, inference health, and overall model performance. Developers can also utilize the new AutoRAG feature, which supports multilingual documents and automated testing for agentic applications through visual pipelines.
Timeline
Red Hat AI 3.5 was released on September 29, 2026.
The Tech Race
This release marks a shift toward hardening AI infrastructure to compete with mature cloud-native environments. By building upon Red Hat OpenShift Virtualization, Red Hat is positioning itself to capture enterprise market share from legacy AI deployment models.
Enterprises can now manage complex AI models on shared infrastructure without compromising security or performance. This improvement allows IT teams to reduce hardware costs while maintaining strict regulatory compliance for their internal applications.
The takeaway
Enterprise AI requires robust observability and safety mechanisms to be viable at scale. Organizations should prioritize tools that enable hardware isolation and automated model testing to ensure long-term deployment success.
Further reading
Learn more about the evolving landscape of enterprise machine learning tools by visiting our Artificial Intelligence section.
Source note: This article includes information reported by IT-Online.
Live Poll
Do you trust that businesses have the necessary controls to deploy AI safely at scale?
