diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2024-12-03 12:59:22 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-12-03 12:59:22 +0100 |
| commit | c5bf589367cd609f4c0ff73a6534bbde7902abe8 (patch) | |
| tree | fa17f82c717d535222c1843fc9fca2d66f4d6ea7 /ggml/include/ggml.h | |
| parent | ccec00939a30aa7762a232ac4dcadba985ef9ee4 (diff) | |
Q5_0_R4 (#121)
* Adding q5_0_r4
We get PP-512(LLaMA-3.1-8B) = 256.7 t/s on a Ryzen-7950X.
We even get TG-128 improvement to 11.7 t/s from 11.1 t/s.
* q5_0_r4: NEON
We get PP-512(LLaMA-3.1-8B) = 99.6 t/s on M2-Max,
up from 71.0 t/s for Q5_0. The difference to mainline llama.cpp
is no longer funny: they get 26.5 t/s for Q5_0.
For TG, we are nor able to fully saturate memory bandwidth
and arrive at 22.1 t/s @ 8 threads. Mainline llama.cpp gets
20.6 t/s for Q5_0.
---------
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, 4 insertions, 0 deletions
diff --git a/ggml/include/ggml.h b/ggml/include/ggml.h index 2358fb76..99c39b9c 100644 --- a/ggml/include/ggml.h +++ b/ggml/include/ggml.h @@ -408,8 +408,10 @@ extern "C" { GGML_TYPE_IQ4_KSS = 146, 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_Q6_0_R4 = 233, GGML_TYPE_COUNT, }; @@ -471,7 +473,9 @@ extern "C" { // GGML_FTYPE_MOSTLY_Q4_0_R4 = 202, // except 1d tensors 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_Q6_0_R4 = 227, // except 1d tensors }; // available tensor operations: |