redhat.com / Page 2

More than 10 years ago, Red Hat began a collaboration with Microsoft rooted in a simple but powerful recognition: enterprise customers need trusted, interoperable platforms that offer true choice.

Explore the world's leading open source AI models in a single library. Compare models based on trusted industry benchmarks and find the perfect fit for your project. Get clear information about model origins and intended use cases. Performance ...

The Friday Five is a weekly Red Hat blog post with 5 of the week's top news items and ideas from or about Red Hat and the technology industry.

Learn how Auto ML and AutoRAG in Red Hat OpenShift AI 3.4 streamline model development and RAG optimization for enterprise-grade automated machine learning. Discover the benefits, use cases, and how they fit into the broader enterprise AI pipeline.

Discover how Red Hat OpenShift AI 3.4 and Red Hat Connectivity Link deliver Models-as-a-Service (MaaS) to centrally govern, secure, and scale enterprise AI model serving.

Learn how Speculators, an open source library, brings speculative decoding from research into production, helping organizations reduce inference latency and cost for large language models (LLMs) without compromising on model quality.

Learn how AMD Instinct MI355X GPU support, intelligent orchestration, and AMD EPYC virtual large language model (vLLM) CPU with ZenDNN backend are transforming enterprise AI with Red Hat AI 3.4. Discover the benefits of high-performance GPU and CPU . ...

For 2027, Red Hat Summit is returning to its roots in the heart of the East Coast's tech corridor--Boston, MA. We invite you to clear your schedule for Red Hat Summit 2027, taking place May 25-27, 2027.

Discover the features and benefits of Red Hat OpenShift Virtualization, a strategic evolution for your virtualization infrastructure. Learn about multi-cluster management, intelligent load balancing, memory oversubscription, advanced networking, ...

In this post, Chris shares that as the rise of AI agents drives token consumption and costs to unsustainable levels, organizations must shift from being passive token consumers to providing their own optimized, self-hosted inference infrastructure. . ...