llama.cpp/ggml
Oliver Simons 7668999518 Merge branch 'master' into gpu-sampling
Let's keep `master's` cumsum implementation for it's likely better AMD
perf and add back pure-CUB-implementation in follow-up commit
2025-12-05 14:41:08 +01:00
..
cmake ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094) 2025-08-07 13:45:41 +02:00
include build : move _WIN32_WINNT definition to headers (#17736) 2025-12-04 07:04:02 +01:00
src Merge branch 'master' into gpu-sampling 2025-12-05 14:41:08 +01:00
.gitignore vulkan : cmake integration (#8119) 2024-07-13 18:12:39 +02:00
CMakeLists.txt build : move _WIN32_WINNT definition to headers (#17736) 2025-12-04 07:04:02 +01:00