Skip to content

PTQ-Level Efficiency, QAT-Level Accuracy | 训练后量化的效率,量化感知训练的精度

Commits on main

add the wrapper for woqgemm (#1965)
Zhenzhong Xu committed 4 months ago
3ab9520
speed up awq ci (#1954)
Weiwei committed 4 months ago
9468e52
5a3ade8
Remove kernels version limit in test (#1945)
chen, suyue committed 4 months ago
142e91b
436b9db
update ARK README (#1922)
Luo Yu committed 4 months ago
f42a042
126793d
fix issue of step3p7 and CI failure (#1929)
Xin He committed 5 months ago
1c4baa0
Update CPU agent pool (#1936)
chen, suyue committed 5 months ago
35650a0
DDP for multiple cards (#1897)
Yi Liu committed 5 months ago
33b7df0
reorder UT for CUDA CI memory leak (#1938)
Xin He committed 5 months ago
0cc47c6
Enhance b70 check (#1937)
Yi Liu committed 5 months ago
e870916
Support Wan2.2 S2V Quantization (#1924)
Wang, Chang committed 5 months ago
ca37879
Fix UT coverage (#1927)
Sun, Xuehao committed 5 months ago
6b31187
df9299d
enable MXFP clamp for model free (#1914)
Xin He committed 5 months ago
50ee58c
fix awq cuda CI (#1912)
Weiwei committed 5 months ago
34790de
fix inplace rotation issue (#1903)
Wenhua Cheng committed 5 months ago
9d5edd5