ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-24 00:19:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

Turn graphs on by default

Browse Source
This commit is contained in:
Iwan Kawrakow
2025-08-15 09:15:48 +03:00
parent 6873fbcbdf
commit 3d5141a44f
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/CMakeLists.txt
View File

@@ -132,7 +132,7 @@ set (GGML_CUDA_MIN_BATCH_OFFLOAD "32" CACHE STRING
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_ALL_QUANTS "ggml: compile all quants for FlashAttention" OFF)
option(GGML_CUDA_USE_GRAPHS "ggml: use CUDA graphs (llama.cpp only)" OFF)
option(GGML_CUDA_USE_GRAPHS "ggml: use CUDA graphs (llama.cpp only)" ON)
option(GGML_IQK_FLASH_ATTENTION "ggml: enable the IQK FlashAttention CPU kernels" ON)
option(GGML_IQK_FA_ALL_QUANTS "ggml: compile all quants for IQK FlashAttention" OFF)