]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 9794052)
Deduplicate q4 quantization functions (#383)
authorStephan Walter <redacted>
Wed, 22 Mar 2023 17:29:06 +0000 (17:29 +0000)
committerGitHub <redacted>
Wed, 22 Mar 2023 17:29:06 +0000 (19:29 +0200)
commit69c92298a9e36dc2363b3bf50452976ce49487b3
tree685e3fc7582eab3173a87d6efd94b3f92070839dtree
parent97940520e8fd49c56bb29b71cc350190b723513fcommit | diff
Deduplicate q4 quantization functions (#383)

* Deduplicate q4 quantization functions

* Use const; add basic test

* Re-enable quantization test

* Disable AVX2 flags in CI

---------

Co-authored-by: Georgi Gerganov <redacted>
.github/workflows/build.yml diff | blob | history
ggml.c diff | blob | history
ggml.h diff | blob | history
tests/CMakeLists.txt diff | blob | history
tests/test-quantize.c [new file with mode: 0644] blob
Packaging of ggml-org/llama.cpp