diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2024-12-17 07:51:11 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-12-17 07:51:11 +0100 |
| commit | d69344f8ea72c6fe6ec16300b939586fa9633e2e (patch) | |
| tree | b8c0efb7322169372543b020360bf0e27549fba5 /ggml/include/ggml.h | |
| parent | 1714e46f137318152370beee16af92991042d7b4 (diff) | |
IQ3_K_R4 (#145)
* iq3_k_r4 WIP
* iq3_k_r4: Zen4
* iq3_k_r4: AVX2
* iq3_k_r4: NEON
* iq3_k_r4: faster matrix x vector multiplication on NEON
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'ggml/include/ggml.h')
| -rw-r--r-- | ggml/include/ggml.h | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/ggml/include/ggml.h b/ggml/include/ggml.h index 7b7fde0d..04831c2f 100644 --- a/ggml/include/ggml.h +++ b/ggml/include/ggml.h @@ -423,6 +423,7 @@ extern "C" { GGML_TYPE_BF16_R16 = 230, GGML_TYPE_Q6_0_R4 = 233, GGML_TYPE_IQ2_BN_R4 = 335, + GGML_TYPE_IQ3_K_R4 = 338, GGML_TYPE_IQ4_K_R4 = 339, GGML_TYPE_Q8_K_R8 = 399, GGML_TYPE_COUNT, @@ -497,6 +498,7 @@ extern "C" { GGML_FTYPE_MOSTLY_BF16_R16 = 224, // except 1d tensors GGML_FTYPE_MOSTLY_Q6_0_R4 = 227, // except 1d tensors GGML_FTYPE_MOSTLY_IQ2_BN_R4 = 329, // except 1d tensors + GGML_FTYPE_MOSTLY_IQ3_K_R4 = 331, // except 1d tensors GGML_FTYPE_MOSTLY_IQ4_K_R4 = 332, // except 1d tensors GGML_FTYPE_MOSTLY_Q8_K_R8 = 399, // except 1d tensors }; |