limnmn/llama.cpp-b9637-Windows-Runtime

1---2license: mit3tags:4  - gguf5  - llama-cpp6  - windows7  - runtime8---9 10# llama.cpp b9637 Windows runtime mirror11 12This repository mirrors an immutable subset of the official13[`ggml-org/llama.cpp` b9637 release](https://github.com/ggml-org/llama.cpp/releases/tag/b9637)14for resilient application installation on Windows x64.15 16Included archives:17 18- Windows x64 CPU19- Windows x64 Vulkan20- Windows x64 CUDA 12.4 executable/runtime files built by llama.cpp21- Windows x64 CUDA 13.3 executable/runtime files built by llama.cpp22 23Every file is unmodified. Verify it against `checksums.sha256` before use.24 25NVIDIA CUDA runtime and cuBLAS redistributable archives are deliberately not26republished here. Applications that select a CUDA build must obtain those27components from the official upstream release under the NVIDIA CUDA EULA, or28fall back to the Vulkan build.29 30This mirror is not affiliated with or endorsed by the llama.cpp project or31NVIDIA.32 33