about.subtitle
The efficiency of modern computing systems is largely determined by how effectively the workload is balanced between the CPU and the GPU. In this section, we explore methods and approaches that allow you to squeeze every drop of performance out of graphics card architecture – from fine-tuning parallel computing to memory management and shader optimization. The materials provided go beyond simple tutorials; they offer a deep dive into how software code interacts with «hardware» at the low level.
Researchers have demonstrated how to fine-tune AI models to replace complex physics simulations – making them faster and cheaper than running calculations from scratch.
AI: Events
Technical context • Research
AMD researchers have demonstrated how an AI agent can iteratively optimize high-performance code without human intervention.
AI: Events
Technical context • Infrastructure
AMD has shared its approach to simplifying the training of recommendation systems on its GPUs – the algorithms that select movies, products, and news for us.
AI: Events
Infrastructure
The new NVIDIA GB300 NVL72 server, paired with the SGLang framework, has demonstrated a 25x performance boost when running language models.
AMD has released JAX-AITER, a library of pre-built, optimized computational blocks for developing large AI models on AMD GPUs using the JAX framework.
AI: Events
Infrastructure
AMD has shared its progress in supporting AI models on its GPUs: from basic compatibility to optimized performance comparable with competitors.
Want to dive deeper into the world
of neuro-creativity?
Be the first to learn about new books, articles, and AI experiments
on our Telegram channel!