diff options
| author | Kawrakow <iwankawrakow@gmail.com> | 2025-07-13 19:14:26 +0200 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-07-13 19:14:26 +0200 |
| commit | 255c22046bcaef41850125be924f3e42e2a65571 (patch) | |
| tree | 06cfaf5c943a444dd013b4e0072369d82c5188b3 | |
| parent | e2b1a5e1fcb3ad55eae03c58c986a21e842ff7a4 (diff) | |
Add iq3_ks to constants.py (#606)
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
| -rw-r--r-- | gguf-py/gguf/constants.py | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py index b3b2bc50..0fe3ed35 100644 --- a/gguf-py/gguf/constants.py +++ b/gguf-py/gguf/constants.py @@ -1320,6 +1320,7 @@ class GGMLQuantizationType(IntEnum): IQ2_KT = 153 IQ3_KT = 154 IQ4_KT = 155 + IQ3_KS = 156 Q4_0_R8 = 202 Q5_0_R4 = 206 Q8_0_R8 = 208 @@ -1535,6 +1536,7 @@ GGML_QUANT_SIZES: dict[GGMLQuantizationType, tuple[int, int]] = { GGMLQuantizationType.IQ2_KT : ( 256, 68), GGMLQuantizationType.IQ3_KT : ( 256, 100), GGMLQuantizationType.IQ4_KT : ( 256, 128), + GGMLQuantizationType.IQ3_KS : ( 256, 102), GGMLQuantizationType.Q4_0_R8 : ( 32, 18), GGMLQuantizationType.Q5_0_R4 : ( 32, 22), GGMLQuantizationType.Q8_0_R8 : ( 32, 34), |