]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 2e9c0bd)
Slight quantization improvement for Q4_K and Q5_K (#5361)
authorKawrakow <redacted>
Tue, 6 Feb 2024 15:28:02 +0000 (17:28 +0200)
committerGitHub <redacted>
Tue, 6 Feb 2024 15:28:02 +0000 (17:28 +0200)
commitf57fadc009cbff741a1961cb7896c47d73978d2c
tree568d11943ed4244323041131c39444cb1376cb8btree
parent2e9c0bd6b301155ce749e162527fc55e9fb5b832commit | diff
Slight quantization improvement for Q4_K and Q5_K (#5361)

* Q4_K: slightly better quantization

* Q5_K: slightly better quantization

---------

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