Your browser does not seem to support JavaScript. As a result, your viewing experience will be diminished, and you have been placed in read-only mode.
Please download a browser that supports JavaScript, or enable it if it's disabled (i.e. NoScript).
用2080 Ti 22G 魔改卡部署Qwen3.8-27B ，无 MTP 26.3 tok/s vs 有 MTP 36.7 tok/s，提升 +39%（比文章的 +20% 还好，因为新版 llama.cpp 的 MTP 实现更成熟）。
