tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-09-06 15:18:01 -04:00
Code Issues Packages Projects Releases Wiki Activity

CUDA: FA support for Deepseek (Ampere or newer) (#13306)

Browse Source 
* CUDA: FA support for Deepseek (Ampere or newer)

* do loop unrolling via C++ template
This commit is contained in:
Johannes Gäßler
2025-05-09 13:34:58 +02:00
committed by GitHub
parent 27ebfcacba
commit 0cf6725e9f
33 changed files with 852 additions and 547 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/src/ggml-cuda/CMakeLists.txt
View File

@@ -118,7 +118,7 @@ if (CUDAToolkit_FOUND)
set(CUDA_CXX_FLAGS "")
set(CUDA_FLAGS -use_fast_math)
set(CUDA_FLAGS -use_fast_math -extended-lambda)
if (CUDAToolkit_VERSION VERSION_GREATER_EQUAL "12.8")
# Options are: