https://blog.landryzetam.net/tags/gpu/Recent content in GPU on Landry Zetam - Tech BlogHugo -- gohugo.ioen-usMon, 11 Aug 2025 00:00:00 +0000https://blog.landryzetam.net/posts/gpustack-mac-clustering/Mon, 11 Aug 2025 00:00:00 +0000https://blog.landryzetam.net/posts/gpustack-mac-clustering/<h2 id="opening-hook">Opening Hook</h2> <p>I&rsquo;ve always loved the idea of running powerful AI workloads without relying on expensive cloud GPUs. Recently, I turned that idea into reality by transforming three Mac Studios into a fully functioning local AI cluster using <strong>GPUStack</strong>. This setup lets me run distributed inference workloads across multiple Macs with Apple Metal acceleration — and it&rsquo;s a lot more affordable and energy-efficient than most people think.</p> <blockquote> <p><strong>Related:</strong> See <a href="https://blog.landryzetam.net/posts/mac-ai-cluster/">Mac AI Cluster: From EXO Vision to GPUStack Reality</a> for the philosophical journey that led to this technical implementation.</p>https://blog.landryzetam.net/posts/nvidia-vs-apple-silicon-architecture/Mon, 11 Aug 2025 00:00:00 +0000https://blog.landryzetam.net/posts/nvidia-vs-apple-silicon-architecture/<p>When I tell people I run AI models on Mac Studios instead of NVIDIA GPUs, I often get puzzled looks. &ldquo;But NVIDIA is the gold standard!&rdquo; they say. And they&rsquo;re not wrong — but they&rsquo;re also not seeing the full picture. The architectural differences between NVIDIA and Apple Silicon aren&rsquo;t just technical minutiae; they fundamentally change how AI models work, how they&rsquo;re trained, and what hardware they run best on.</p>