llama.cpp
AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring
#1099

Merged

AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring #1099

ggerganov merged 3 commits into ggml-org:master from sw:q43-avx

slaren approved these changes on 2023-04-21

AVX2 optimization for vec_dot_q4_3_q8_0 and refactoring

63d8dff4

sw force pushed from a32f5813 to 63d8dff4 2 years ago

sw marked this pull request as draft 2 years ago

finish AVX vectorization of quantize_row_q8_0

535ea470

sw marked this pull request as ready for review 2 years ago

pubby commented on 2023-04-21

Rename hsum_int_8 to hsum_i32_8

3a5958bd

ggerganov approved these changes on 2023-04-22

ggerganov merged c5aa5e57 into master 2 years ago

sw deleted the q43-avx branch 2 years ago

Reviewers

ggerganov

slaren

pubby

Assignees

No one assigned

Labels

None yet

Milestone

No milestone