llama.cpp/ggml
uvos e721c05c93
HIP/CUDA: set the paramerter value in maintain_cuda_graph instead of replaceing it. (#12209)
This avoids conflict with internal cuda/hip runtimes memory managment behavior.
2025-03-06 08:20:52 +01:00
..
cmake cmake: Fix ggml backend dependencies and installation (#11818) 2025-02-27 09:42:48 +02:00
include ggml-cpu: Faster IQ1 mul_mat_vec on AVX2 using BMI2 instructions (#12154) 2025-03-06 02:26:10 +01:00
src HIP/CUDA: set the paramerter value in maintain_cuda_graph instead of replaceing it. (#12209) 2025-03-06 08:20:52 +01:00
.gitignore vulkan : cmake integration (#8119) 2024-07-13 18:12:39 +02:00
CMakeLists.txt ggml-cpu: Faster IQ1 mul_mat_vec on AVX2 using BMI2 instructions (#12154) 2025-03-06 02:26:10 +01:00