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

refine readme (#798)
Wenhua Cheng committed 1 year ago
bc045f9
add fp8 export format check (#779)
Heng Guo committed 1 year ago
5ead4c6
align formats packing device to API (#795)
Weiwei committed 1 year ago
dc81a87
Support auto device mapping (#781)
Tang Kaihui committed 1 year ago
d7d2efa
torch backend bugfix and speedup ut (#793)
Weiwei committed 1 year ago
4c597de
add limit arg for eval (#764)
Heng Guo committed 1 year ago
add749a
refine nvfp4 algorithm (#790)
Wenhua Cheng committed 1 year ago
536810e
5f87a06
refine nvfp code, typofix (#777)
Weiwei committed 1 year ago
fced2ba
hot fix for nfp4 scheme (#784)
Wenhua Cheng committed 1 year ago
7e52af2
Speedup FP4 packing (#766)
Yi Liu committed 1 year ago
253b48f
gguf format supports for fp8 model (#778)
Heng Guo committed 1 year ago
572aad1
fix bug of auto_round exporting (#772)
Heng Guo committed 1 year ago
d01704d
fix set_layer_config bug (#768)
Wenhua Cheng committed 1 year ago
bf58cf9
fix gguf bug by int zp (#771)
Heng Guo committed 1 year ago
f01e804
fix gguf ut bug (#767)
Heng Guo committed 1 year ago
48ab0db
Revert "Speedup FP4 packing (#760)" (#763)
Yi Liu committed 1 year ago
ec76f61
5e40699
Speedup FP4 packing (#760)
Yi Liu committed 1 year ago
3648639
fix some torch compile warnings (#755)
Wenhua Cheng committed 1 year ago
1f1d0a9
fix alg_ext moe and model str input bug (#751)
Wenhua Cheng committed 1 year ago
8f84bcb