llama.cpp
k_quants : add AVX support to dot functions with QK_K as 64
#2339
Merged

k_quants : add AVX support to dot functions with QK_K as 64 #2339

ggerganov merged 6 commits into ggml-org:master from katsu560:add-avx-qkk64
katsu560
katsu560 add AVX to ggml_vec_dot_q2_K_q8_K()
6afbb11c
katsu560 add AVX to ggml_vec_dot_q3_K_q8_K()
2024d40a
katsu560 add AVX to ggml_vec_dot_q4_K_q8_K()
56df218f
katsu560 add AVX to ggml_vec_dot_q5_K_q8_K()
dfd2e0e5
katsu560 add AVX to ggml_vec_dot_q6_K_q8_K()
4775602c
katsu560 refactor AVX code in ggml_vec_dot_q6_K_q8_K()
5f98fc2f
netrunnereve
ikawrakow
katsu560
ikawrakow
ikawrakow approved these changes on 2023-07-24
katsu560
ggerganov ggerganov merged be2301bc into master 2 years ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone