]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: e64f5b5)
metal : Q3_K speedup (#2995)
authorKawrakow <redacted>
Fri, 8 Sep 2023 16:01:04 +0000 (18:01 +0200)
committerGitHub <redacted>
Fri, 8 Sep 2023 16:01:04 +0000 (19:01 +0300)
commitba7ffbb2517ff8cf4c689f94a9ad866f3ee71225
treeb30e2122c9d3d637a9f6145089a92e574b535c6atree
parente64f5b55783e910d8287363895d652b4bea6527acommit | diff
metal : Q3_K speedup (#2995)

* Slightly faster Q3_K and Q5_K on metal

* Another Q3_K speedup on metal

Combined with previous commit, we are now +9.6% for TG.
PP is not affected as this happens via the matrix multiplication
templates.

* Slowly progressing on Q3_K on metal

We are now 13% faster than master

* nother small improvement for Q3_K on metal

---------

Co-authored-by: Iwan Kawrakow <redacted>
ggml-metal.metal diff | blob | history
Packaging of ggml-org/llama.cpp