{"type": "article", "title": "Building a single-kernel, latency-optimized LLM inference engine on AMD MI300X GPUs", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/building-a-single-kernel-latency-optimized-llm-inference-engine-on-amd-mi300x", "original_source": "https://blog.kog.ai/building-a-single-kernel-latency-optimized-llm-inference-engine-on-amd-mi300x-gpus/", "published": "2026-05-28T16:18:00+00:00", "accessed": "2026-07-14", "id": "building-a-single-kernel-latency-optimized-llm-inference-engine-on-amd-mi300x"}