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

update readme for sglang support (#953)
Weiwei committed 1 year ago
824a21f
refactor utils file (#943)
Heng Guo committed 1 year ago
f1b5c72
update readme (#949)
Wenhua Cheng committed 1 year ago
959ac67
Add diffusion README (#923)
Wang, Mengni committed 1 year ago
97512a4
support attention mask in user's dataset (#930)
Wenhua Cheng committed 1 year ago
dfbbba4
Add commit hash to version (#941)
Sun, Xuehao committed 1 year ago
9052e31
Add compatibility test (#918)
Sun, Xuehao committed 1 year ago
045a27d
remove low_cpu_mem (#934)
Heng Guo committed 1 year ago
1355311
90a3e86
Fix Flux tuning issue (#936)
Wang, Mengni committed 1 year ago
22bd6b0
remove numpy<2.0 limitation (#921)
Xin He committed 1 year ago
a00fcc6
optimize rtn for int woq (#924)
Wenhua Cheng committed 1 year ago
da55e27
5bb16b0
Extend mxfp loading dtypes (#916)
Yi Liu committed 1 year ago
0164077
add numba for cpu lib (#919)
Yi Liu committed 1 year ago
41986b5
fix llama 4 ut (#896)
Heng Guo committed 1 year ago
b738ea0
support disable_opt_rtn in auto-scheme (#913)
Wenhua Cheng committed 1 year ago
5de9fbc
Revert "Extend mxfp loading dtypes (#907)" (#915)
Wenhua Cheng committed 1 year ago
46812de
fix torch compile issue in AutoScheme (#909)
Wenhua Cheng committed 1 year ago
1d91207
block dataset logger info (#908)
Heng Guo committed 1 year ago
fe62213
Extend mxfp loading dtypes (#907)
Yi Liu committed 1 year ago
0c2619c
remove numba from requirments (#905)
Yi Liu committed 1 year ago
56f469e
try to fix gguf issue (#886)
Wenhua Cheng committed 1 year ago
6c8974b
support automatic mixed bits assignment (#851)
Wenhua Cheng committed 1 year ago
0f64ed0
fix nvfp act quantization bug (#891)
Weiwei committed 1 year ago
fbc41c8