news
Red Hat Leftovers
-
Red Hat ☛ Deploying OpenShift hosted clusters on bare metal
Deploying Red Hat OpenShift hosted clusters on bare metal (often referred to as Hypershift) is a game-changer for infrastructure efficiency. By decoupling control planes from worker nodes, you can dramatically reduce hardware costs and dedicate more physical resources to running the virtual machines that matter.
As with any powerful technology, the official documentation provides the “what,” but the “how” and “why” are often discovered in the field. The rest is a collection of practical knowledge: the non-obvious truths and critical procedures you only learn through hands-on deployment, troubleshooting, and long hours in the command line. This is my guide to the five most surprising and impactful lessons learned from the field; things I wish someone had told me before I started.
-
Red Hat ☛ Get started with language model post-training using Training Hub
Open source language models can deliver incredible performance because of their smaller size and efficient cost, but they might not deliver the required results for more targeted use cases. There is, however, one more key benefit that open source models have over their frontier model counterparts: customizability. Often, you can fit an open source model to your specific use case via post-training, which offers better and cheaper performance than even frontier models can provide. while running securely and safely for private data, offline platforms, and so on.
-
Red Hat ☛ Speculators: Standardized, production-ready speculative decoding
Large language models (LLMs) are powerful, but slow. Every token requires a forward pass through billions of parameters, which quickly adds up at scale. Speculative decoding flips the script: [...]
-
Red Hat ☛ Defining success: Evaluation metrics and data augmentation for oversaturation detection
Oversaturation is a sneaky problem that wastes time, money, and costly GPU cycles when benchmarking large language models (LLMs). In Reduce LLM benchmarking costs with oversaturation detection, we established what oversaturation is and explained why oversaturation detection (OSD) is crucial for controlling our LLM benchmarking budgets.
Now, we're moving from the problem to the solution. But how do you teach a machine to spot a condition that is difficult to even define? Here's how we built the algorithm.
Our goal: Don't waste money
Our goal is simple, but it has two conflicting parts: [...]
-
Unicorn Media ☛ Red Hat Project Hummingbird For CVE-Free, Ready-to-Deploy Container Images
Project Hummingbird aims to remove the worry and streamline the deployment of secure container images for cloud-native teams.