llama.cpp/ggml
Aman Gupta 38355c6c8e
CUDA: use registers instead of smem in topk-moe (#16647)
Uses the technique used in the vulkan PR #16641. Neat trick!
2025-10-18 11:52:53 +02:00
..
cmake ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094) 2025-08-07 13:45:41 +02:00
include rpc : report actual free memory (#16616) 2025-10-17 18:02:52 +03:00
src CUDA: use registers instead of smem in topk-moe (#16647) 2025-10-18 11:52:53 +02:00
.gitignore vulkan : cmake integration (#8119) 2024-07-13 18:12:39 +02:00
CMakeLists.txt ggml webgpu: profiling, CI updates, reworking of command submission (#16452) 2025-10-07 13:48:56 -07:00