llama.cpp 发布: b11064

来源摘要

metal : support arbitrary hc in dsv4_hc_pre (#29169) the dsv4_hc_pre kernels hardcoded hc = 4 via a constexpr used with simd_shuffle, so the op was rejected by supports_op for any other hc and fell back to CPU. Kimi-K3 uses dsv4_hc_pre with hc equal to the number of banked checkpoints in the cross-layer residual stack, which grows with the layer index. pass n_hc as a function constant (FC_DSV4_HC) with per-n_hc pipeline variants, and loop over it in both pre kernels with direct loads add test-backend-ops cases for hc = 1, 2, 3, 5, 8 and 65, gated and not gated Assisted-by: pi:llama.cpp/Qwen3.8-27B **Website:** - **Attestations:** - **macOS/iOS:** - [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-macos-arm64.tar.gz) - macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780) - [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-macos-x64.tar.gz) - [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-xcframework.zip) **Linux:** - [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-ubuntu-x64.tar.gz) - [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-ubuntu-arm64.tar.gz) - [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-ubuntu-s390x.tar.gz) - [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-ubuntu-vulkan-x64.tar.gz) - [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b11064/llama-b11064-bin-ubuntu-vulkan-arm64.tar.gz) - [Ubuntu x64 (CUDA 12)](htt