diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2024-12-04 15:20:07 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-12-04 15:20:07 +0100 |
| commit | f64de08203aaee95ca755336de3e1db85d990198 (patch) | |
| tree | 9af01056e0b304ee5df5792f25d82066931eb4d6 /ggml/include/ggml.h | |
| parent | f1f4eb988fe5ee969100cd0d3782fd7460d13949 (diff) | |
IQ4_XS_R4 (#123)
* Adding iq4_xs_r4
This is a 1st working version on Zen4.
We get PP-512(LLaMA-3.1-8B) = 226 t/s, so 16% slower
than iq4_nl_x4.
* iq4_xs_r4: WIP
* iq4_xs_r4: Use AVX2 version for matrix x vector on Zen4
* iq4_xs_r4: NEON
We get PP-512(LLaMA-3.1-8B) = 115.6 t/s on M2-Max,
up from 68.2 t/s for iq4_xs!
* DRY
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'ggml/include/ggml.h')
| -rw-r--r-- | ggml/include/ggml.h | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/ggml/include/ggml.h b/ggml/include/ggml.h index 99c39b9c..09f92eb9 100644 --- a/ggml/include/ggml.h +++ b/ggml/include/ggml.h @@ -410,7 +410,8 @@ extern "C" { GGML_TYPE_Q4_0_R4 = 202, GGML_TYPE_Q5_0_R4 = 206, GGML_TYPE_Q8_0_R4 = 208, - GGML_TYPE_IQ4_NL_X4 = 220, + GGML_TYPE_IQ4_NL_X4 = 220, // TODO: rename GGML_TYPE_IQ4_NL_X4 to GGML_TYPE_IQ4_NL_R4 + GGML_TYPE_IQ4_XS_R4 = 223, GGML_TYPE_Q6_0_R4 = 233, GGML_TYPE_COUNT, }; @@ -475,6 +476,7 @@ extern "C" { GGML_FTYPE_MOSTLY_Q8_0_R4 = 207, // except 1d tensors GGML_FTYPE_MOSTLY_Q5_0_R4 = 208, // except 1d tensors GGML_FTYPE_MOSTLY_IQ4_NL_X4 = 219, // except 1d tensors + GGML_FTYPE_MOSTLY_IQ4_XS_R4 = 222, // except 1d tensors GGML_FTYPE_MOSTLY_Q6_0_R4 = 227, // except 1d tensors }; |