ikawrakow/ik_llama.cpp
1
0
Fork 0
mirror of https://github.com/ikawrakow/ik_llama.cpp.git synced 2026-04-30 19:31:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add iq3_ks to constants.py (#606)

Browse Source 
Co-authored-by: Iwan Kawrakow <iwan.kawrakow@gmail.com>
This commit is contained in:
Kawrakow
2025-07-13 19:14:26 +02:00
committed by GitHub
parent e2cf466eaa
commit 4f56069442
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
gguf-py/gguf/constants.py
View File

@@ -1320,6 +1320,7 @@ class GGMLQuantizationType(IntEnum):
IQ2_KT = 153 IQ2_KT = 153
IQ3_KT = 154 IQ3_KT = 154
IQ4_KT = 155 IQ4_KT = 155
IQ3_KS = 156
Q4_0_R8 = 202 Q4_0_R8 = 202
Q5_0_R4 = 206 Q5_0_R4 = 206
Q8_0_R8 = 208 Q8_0_R8 = 208
@@ -1535,6 +1536,7 @@ GGML_QUANT_SIZES: dict[GGMLQuantizationType, tuple[int, int]] = {
GGMLQuantizationType.IQ2_KT : ( 256, 68), GGMLQuantizationType.IQ2_KT : ( 256, 68),
GGMLQuantizationType.IQ3_KT : ( 256, 100), GGMLQuantizationType.IQ3_KT : ( 256, 100),
GGMLQuantizationType.IQ4_KT : ( 256, 128), GGMLQuantizationType.IQ4_KT : ( 256, 128),
GGMLQuantizationType.IQ3_KS : ( 256, 102),
GGMLQuantizationType.Q4_0_R8 : ( 32, 18), GGMLQuantizationType.Q4_0_R8 : ( 32, 18),
GGMLQuantizationType.Q5_0_R4 : ( 32, 22), GGMLQuantizationType.Q5_0_R4 : ( 32, 22),
GGMLQuantizationType.Q8_0_R8 : ( 32, 34), GGMLQuantizationType.Q8_0_R8 : ( 32, 34),