news
Fedora and Red Hat Leftovers
-
Red Hat ☛ Enable 3.5 times faster vision language models with quantization
Performant deployments: With vLLM, up to 3.5 times faster throughput scenarios and 3.2 times more requests per second for server scenarios.
Vision-language models (VLMs), such as the Pixtral and Qwen-VL series, are trained to generate text from image and text inputs. With the expanded input types and the performance of large language models, they enable accurate and promising new use cases such as content moderation, image captioning and tagging, visual question answering, and document extraction/analysis, among others. The extra modality, though, means that VLMs are even more computationally demanding, requiring more processing power and memory than the already demanding language-only architectures.
-
Red Hat ☛ How to set up OpenShift confidential clusters on Azure [Ed: IBM Red Hat is trying to sell back doors for Microsoft and endless, limitless surveillance under the false guise of "confidential"]
-
Michel Alexandre Salim ☛ Spreading Beefy Miracle love
Many long-time Fedora contributors and users have fond memories of Beefy Miracle, the Fedora 17 code name, and its mascot. Major Haydn has a great writeup about it, with instructions on switching back to its Plymouth boot splash on your Fedora machine.
But… why limit one’s chance at “accepting the truth and beauty” (sic) to Fedora users (or, strictly speaking, to one’s Fedora machines)? This simply would not do. This sort of fun does not deserved to be contained.