8000 Q8: use int8_t, AVX/AVX2 optimizations by sw · Pull Request #972 · ggml-org/llama.cpp · GitHub
[go: up one dir, main page]
Skip to content

Q8: use int8_t, AVX/AVX2 optimizations#972

Merged
ggerganov merged 2 commits intoggml-org:q8_0from
sw:mulmat-q8
Apr 14, 2023
Merged

Q8: use int8_t, AVX/AVX2 optimizations#972
ggerganov merged 2 commits intoggml-org:q8_0from
sw:mulmat-q8

Commits

Commits on Apr 14, 2023
0