Red Hat, the world’s leading open source solutions provider, announced Red Hat AI 3, a major evolution of its enterprise AI platform. The platform integrates Red Hat AI Inference Server, RHEL AI, and Red Hat OpenShift AI, enabling organizations to simplify high-performance AI inference at scale and move workloads from proofs-of-concept to production more efficiently.
Red Hat AI 3 simplifies distributed AI inference across hybrid and multi-vendor environments.
Supports any AI model on any hardware accelerator, from datacenters to cloud and edge.
Enables enterprise-scale production of large language models (LLMs) with llm-d and vLLM.
Unified platform supports collaborative AI workflows for IT and AI engineers.
Provides foundation for agentic AI systems and next-generation AI applications.
Partners include AMD, NVIDIA, and ARSAT for accelerated, scalable AI workloads.
Red Hat AI 3 emphasizes the shift from AI model training to inference, delivering production-ready capabilities. Key innovations include:
llm-d: Distributed, Kubernetes-native LLM inference for intelligent scheduling and disaggregated serving.
vLLM integration: High-performance, scalable serving system for large AI models, including Mixture-of-Experts (MoE).
Well-lit Paths: Prescriptive deployment guidance to simplify model deployment on hybrid cloud environments.
Cross-platform support: Enables LLM inference across NVIDIA and AMD accelerators.
Red Hat AI 3 unifies workflows for IT and AI engineers, providing tools to scale from experimentation to production:
Model as a Service (MaaS): IT teams can centrally manage and deliver models on demand.
AI Hub: Central catalog for exploring, deploying, and monitoring foundational AI models.
Gen AI Studio: Hands-on environment for prototyping generative AI applications with interactive model playgrounds.
Curated validated models, including gpt-oss, DeepSeek-R1, Whisper, and Voxtral Mini.
Red Hat AI 3 enables scalable agentic AI systems through:
Unified API layer based on Llama Stack for OpenAI-compatible LLM integration.
Model Context Protocol (MCP) adoption for seamless AI model interoperability.
Modular toolkit for model customization, including InstructLab, Docling, synthetic data generation, and evaluation hub for model validation.
“With Red Hat AI 3, we are providing an enterprise-grade, open source platform that minimizes complexity, cost and control challenges. By bringing distributed inference with llm-d and a foundation for agentic AI, we enable IT teams to operationalize next-generation AI on their own terms.” — Joe Fernandes, VP & GM, AI Business Unit, Red Hat“As Red Hat brings distributed AI inference into production, AMD provides the high-performance foundation with EPYC™ processors and Instinct™ GPUs to deliver scalable enterprise AI.” — Dan McNamara, SVP & GM, Server and Enterprise AI, AMD“Red Hat OpenShift AI enabled ARSAT to move from concept to production in just 45 days, improving service while preserving data sovereignty.” — Mariano Greco, CEO, ARSAT“Scalable, high-performance inference is key to the next wave of generative and agentic AI. Red Hat AI 3 empowers teams to move swiftly from experimentation to production workloads at scale.” — Ujval Kapasi, VP, Engineering AI Frameworks, NVIDIA
Red Hat is the open hybrid cloud technology leader, delivering a trusted, consistent foundation for IT innovation and AI applications. Its portfolio includes cloud, developer, AI, Linux, automation, and application platform technologies, enabling enterprises to deploy applications anywhere—from datacenter to edge. Red Hat invests in open ecosystems, partnering with customers to build, connect, automate, secure, and manage IT environments, supported by consulting services and award-winning training.
TechIntelPro empowers B2B technology decision-makers with the latest industry trends, deep insights, leadership perspectives and actionable intelligence to drive both organizational and professional milestones—and thrive on the tech pulse.