leejet bbac6a26b2 ggml: fix cuda kernel launch configuration for k_compute_batched_ptrs to support large batch (#16744) hai 3 meses
..
cmake 9a96389544 ggml: Skip backend library linking code when GGML_BACKEND_DL=ON (#15094) hai 5 meses
include 63d2fc46e1 Add experimental ggml-hexagon backend for the Hexagon NPU (#16547) hai 3 meses
src bbac6a26b2 ggml: fix cuda kernel launch configuration for k_compute_batched_ptrs to support large batch (#16744) hai 3 meses
.gitignore 17eb6aa8a9 vulkan : cmake integration (#8119) hai 1 ano
CMakeLists.txt 63d2fc46e1 Add experimental ggml-hexagon backend for the Hexagon NPU (#16547) hai 3 meses