Allen Kuo (kwyshell)·3d agoAnother Route to vLLM-Style Serving, Part 2: CPU, Vulkan, and GGUF Across PlatformsHandwritten and Triton-CPU kernels, Vulkan tiling and data reuse, GGUF format costs, and the gaps that remain
Allen Kuo (kwyshell)·3d agovLLM 式 serving 的另一條路（下）：CPU、Vulkan 與 GGUF，走向多平台手寫與 Triton-CPU 的混合 kernel、Vulkan 的分塊與資料重用、GGUF 的格式代價，以及還沒追上的地方
Samira·Oct 23𝗖𝗮𝗰𝗵𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝘀𝘁𝗱::𝘄𝗲𝗮𝗸_𝗽𝘁𝗿 𝗶𝗻 𝐂++🎯𝗖𝗮𝗰𝗵𝗶𝗻𝗴 𝘄𝗶𝘁𝗵 𝘀𝘁𝗱::𝘄𝗲𝗮𝗸_𝗽𝘁𝗿 𝗶𝗻 𝐂++ There are a few potential use cases for std::weak_ptr: ✅Caching ✅Observer lists…
Chekwube Utomi·Oct 21Memory management in C++ explained to dullardsOne of the features that distinguishes C++ from languages like Python, Java, and JavaScript is its approach to memory management. While…
Todd Bandrowsky·Oct 20Corona, a C++ object database and stack for Windows NativeCorona is an experiment in pushing abstraction to ever higher levels to drive down costs of development while improving correctness of…
