Red Hat has announced the launch of Red Hat AI Enterprise, an integrated platform designed to deploy and manage artificial intelligence (AI) models, agents, and applications across hybrid cloud environments. The new product is part of the company’s broader AI portfolio, which includes Red Hat AI Inference Server, Red Hat OpenShift AI, and Red Hat Enterprise Linux AI. Alongside this release, the company introduced Red Hat AI 3.3 with updates aimed at improving performance and expanding features across its suite of AI tools.
The introduction of Red Hat AI Enterprise aims to address challenges organizations face in moving from experimental projects to fully operationalized and governed AI systems. Many enterprises have struggled to move beyond pilot phases due to fragmented tools and inconsistent infrastructure. The new platform seeks to unify model and application lifecycles so that IT teams can manage AI as a standardized system within their organizations.
Joe Fernandes, vice president and general manager of the AI Business Unit at Red Hat, said: “For AI to deliver true business value, it must be operationalized as a core component of the enterprise software stack, not as a standalone silo. Red Hat AI Enterprise is designed to bridge the gap between infrastructure and innovation by providing a unified metal to agent platform. By integrating advanced tuning and agentic capabilities with the industry-leading foundation of Red Hat Enterprise Linux and Red Hat OpenShift, we are providing the complete stack – from the GPU-accelerated hardware to the models and agents that drive business logic. Additionally, with Red Hat AI 3.3 organizations can move beyond fragmented pilots to governed, repeatable and high-performance AI operations across the hybrid cloud.”
Red Hat states that its new solution provides core functions such as high-performance inference for generative models using vLLM inference engine technology; lifecycle management for governance; flexibility in deployment on various hardware platforms; support for NVIDIA’s Blackwell Ultra processors; AMD MI325X accelerators; and previewed support for Intel CPUs for cost-effective small language model inference.
A notable feature is integration with NVIDIA through co-engineering efforts on the Red Hat AI Factory, combining capabilities from both companies’ enterprise platforms.
The latest version also introduces Models-as-a-Service (MaaS) in technology preview mode—allowing internal users self-service access via an API gateway—and enhancements such as expanded model options available through OpenShift’s catalog. These include compressed versions of popular models like Mistral-Large-3 and Nemotron-Nano.
Additional improvements focus on security through a new Python Index repository delivering hardened versions of essential tools; real-time telemetry for observability into workload health; integrated NeMo Guardrails previewed for operational safety enforcement; improved Whisper speech processing speeds; geospatial data support; enhanced tool calling functionality; pooled GPU resource management with automatic checkpointing for training jobs.
Red Hat continues positioning itself as a provider focused on making enterprise-level adoption of advanced machine learning more accessible by offering end-to-end solutions built atop widely used open-source technologies like Linux and Kubernetes.
More information about these developments can be found in the official blog post. Details about joint work with NVIDIA are available via this link. For those interested in further updates or technical roadmaps related to these products, additional sessions are scheduled by the company.
