llama.cpp
AVX Q4_0 and Q8_0 sgemm
#6891
Merged

AVX Q4_0 and Q8_0 sgemm #6891

ggerganov merged 12 commits into ggml-org:master from sgemm-avx
netrunnereve
netrunnereve basic avx implementation
86d1d846
netrunnereve style
257391aa
netrunnereve combine denibble with load
9facb0f0
netrunnereve reduce 256 to 128 (and back!) conversions
dee9566d
netrunnereve sse load
063a31f7
netrunnereve Merge branch 'ggerganov:master' into sgemm-avx
e97c0fdb
netrunnereve Update sgemm.cpp
fb80f13c
cebtenzzre
netrunnereve Merge branch 'ggerganov:master' into sgemm-avx
330b3bc5
jart
jart commented on 2024-04-28
jart
jart approved these changes on 2024-04-29
jart
netrunnereve merge
8916954a
netrunnereve oops
ae0b5ea7
netrunnereve Merge branch 'ggerganov:master' into sgemm-avx
8af6f9c5
netrunnereve Merge branch 'ggerganov:master' into sgemm-avx
f87fc83c
netrunnereve
ggerganov
ggerganov approved these changes on 2024-05-08
ggerganov ggerganov merged 465263d0 into master 1 year ago
netrunnereve netrunnereve deleted the sgemm-avx branch 1 year ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone