{"type": "article", "title": "Fusing a 27B ternary LLM's whole decode step into one CUDA kernel", "publisher": "Web Pulse", "url": "https://wpnews.pro/news/fusing-a-27b-ternary-llm-s-whole-decode-step-into-one-cuda-kernel", "original_source": "https://twitter.com/Akashi203/status/2077552491567157733", "published": "2026-07-16T00:39:18+00:00", "accessed": "2026-07-16", "id": "fusing-a-27b-ternary-llm-s-whole-decode-step-into-one-cuda-kernel"}