Home Explore Help
Sign In
cturan
/
llama.cpp
mirror of https://github.com/cturan/llama.cpp
1
0
Fork 0
Files Issues 0 Wiki
Browse Source

CUDA: deduplicate mmq code (#7397)

Johannes Gäßler 1 year ago
parent
d7e852c1bc
commit
d8ee902227
1 changed files with 224 additions and 956 deletions
Split View Show Diff Stats
  1. 224 956
      ggml-cuda/mmq.cu

File diff suppressed because it is too large
+ 224 - 956
ggml-cuda/mmq.cu


Some files were not shown because too many files changed in this diff