diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2025-07-02 09:27:47 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-07-02 09:27:47 +0200 |
| commit | 3248a3599275c3d2a98bc41d512e1d3ba66e32ba (patch) | |
| tree | 14bdaae22a6db922aa9eb0a7131141da81dfd251 /ggml/src/ggml-quants.c | |
| parent | 46f2e5d249568b6395d17a80feb2ba9ab48156e5 (diff) | |
Adding IQ3_KS quants (#566)
* iq3_ks: basics
* iq3_ks: CUDA dequantize
* iq3_ks: CUDA mmvq
* iq3_ks: mmq
* iq3_ks: faster mmq
* iq3_ks: Zen4
* iq3_ks: AVX2 convert to q8_k_r8
This gives usPP-512 = 360 t/s.
* iq3_ks: AVX2 GEMM/GEMV
* iq3_ks: NEON GEMM/GEMV
* iq3_ks: NEON convert to q8_k_r8
This gives us PP-512 = 164 t/s.
* iq3_ks: Metal dequantize
* iq3_ks: Metal gemv - pathetic performance
---------
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
Diffstat (limited to 'ggml/src/ggml-quants.c')
| -rw-r--r-- | ggml/src/ggml-quants.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/ggml/src/ggml-quants.c b/ggml/src/ggml-quants.c index 220c0c99..96242727 100644 --- a/ggml/src/ggml-quants.c +++ b/ggml/src/ggml-quants.c @@ -15425,6 +15425,7 @@ bool ggml_validate_row_data(enum ggml_type type, const void * data, size_t nbyte case GGML_TYPE_IQ3_KT: break; case GGML_TYPE_IQ4_KT: break; case GGML_TYPE_IQ3_K: break; + case GGML_TYPE_IQ3_KS: break; case GGML_TYPE_IQ4_K: break; case GGML_TYPE_IQ5_K: break; case GGML_TYPE_IQ6_K: break; |