]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: e562b97)
Make IQ1_M work for QK_K = 64 (#6327)
authorKawrakow <redacted>
Wed, 27 Mar 2024 07:44:27 +0000 (08:44 +0100)
committerGitHub <redacted>
Wed, 27 Mar 2024 07:44:27 +0000 (08:44 +0100)
commitcbc83436197cde617cad696e665879c20df77daa
tree303401b02237154b13d8cb9cbb32f0df8dcacdbbtree
parente562b9714b9b3e242361a7f74bbbeb00f6bd99accommit | diff
Make IQ1_M work for QK_K = 64 (#6327)

* iq1_m: make it work for QK_K = 64 (WIP)

* iq1_m: make it work for QK_K = 64 (scalar and AVX2)

* iq1_m: QK_K = 64 seems to work on Metal and ARM_NEON

---------

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