llama.cpp b12128：修复 TOPK_MOE 测试 CI 抖动

AI 摘要

llama.cpp 发布 b12128 版本，主要修复 TOPK_MOE 测试因路由边界近似并列导致的 CI 不稳定问题。修复方式改为通过反演门控函数构造 logits，使各后端上的专家与分组决策间隔一个网格步长，避免浮点 ulp 差异引发误判。同时该版本提供 macOS、Linux、Windows、Android 等多平台预编译包，覆盖 CUDA、Vulkan、ROCm、SYCL、OpenVINO 等后端。

正文节选

<details open> tests: fix CI flakes of TOPK_MOE caused by near ties (#31426) * tests: fix CI flakes of TOPK_MOE caused by near ties With uniform random logits, two experts sometimes land within one f32 ulp of each other at a routing boundary, mostly with sqrt_softplus gating whose output range is narrow. A backend whose gating differs by a few ulps from the CPU then picks the other expert and the test fails. Draw each row of logits again until the group selection and the final top-k both hav

发布时间：2026-11-05 22:03

抓取时间：2026-11-05 22:21

来源机构：ggml-org