diff options
| author | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-06-05 11:27:55 +0300 |
|---|---|---|
| committer | Iwan Kawrakow <iwan.kawrakow@gmail.com> | 2024-06-22 12:02:49 +0300 |
| commit | 47ae12bbec4eedf842ac067534e94bb1214ace73 (patch) | |
| tree | 990d777a70cae29e25344ee0ebd26b21395b60fa /ggml.c | |
| parent | dc96d5484fc0d7626181a6f06dd4f6c0e912cb68 (diff) | |
iqk_mul_mat: experimenting with zen4 (iq2_xxs)
Observing again the wierdness of performance drop
in a quant because of a change in another quant.
After I added FANCY_SIMD implementations for
ia3_s, iq2_s and iq2_xs, I'm observing that
iq2_xxs PP performance dropped to 130 t/s from 139 t/s.
Adding FANCY_SIMD implementation for applying the signs
brings it back to 137 t/s and gives a small boost
for TG as well (23.4 vs 23.0 t/s)
Diffstat (limited to 'ggml.c')
0 files changed, 0 insertions, 0 deletions