ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

CUDA: app option to compile without FlashAttention (#12025)

Browse source
This commit is contained in:
Johannes Gäßler 2025-02-22 20:44:34 +01:00 • committed by GitHub
parent 36c258ee92
commit a28e0d5eb1
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
13 changed files with 46 additions and 31 deletions
Download patch file Download diff file Expand all files Collapse all files

1
ggml/CMakeLists.txt
View file

@ -151,6 +151,7 @@ set (GGML_CUDA_PEER_MAX_BATCH_SIZE "128" CACHE STRING
"ggml: max. batch size for using peer access")
option(GGML_CUDA_NO_PEER_COPY "ggml: do not use peer to peer copies" OFF)
option(GGML_CUDA_NO_VMM "ggml: do not try to use CUDA VMM" OFF)
option(GGML_CUDA_FA "ggml: compile ggml FlashAttention CUDA kernels" ON)
option(GGML_CUDA_FA_ALL_QUANTS "ggml: compile all quants for FlashAttention" OFF)
option(GGML_CUDA_GRAPHS "ggml: use CUDA graphs (llama.cpp only)" ${GGML_CUDA_GRAPHS_DEFAULT})