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 code (#749)
Wenhua Cheng committed 1 year ago
e9b2b83
Add Static FP8 KV Support (#737)
Yi Liu committed 1 year ago
fe1db09
improved algorithm for int2 (#748)
Wenhua Cheng committed 1 year ago
64e7611
fix bug of loading fp8 model (#747)
Heng Guo committed 1 year ago
0029322
support llama4 quant (#744)
Wang, Mengni committed 1 year ago
2df63f2
fix UT check for int zp (#745)
Weiwei committed 1 year ago
1f79db9
Optimize pack zeros for int sym (#743)
Weiwei committed 1 year ago
8888dda
support auto-round-mllm command (#742)
Heng Guo committed 1 year ago
cd980c7
[WIP] support for glm4.5 gguf (#735)
Heng Guo committed 1 year ago
4f740cc
Enable mxfp exporting (#649)
Weiwei committed 1 year ago
64b8afe
fix gguf mistral_common import (#736)
Heng Guo committed 1 year ago
92d31fa
fix bug for exporting afp8 fake format (#731)
Heng Guo committed 1 year ago
978cac1
de2b48a
73ecd7f
fix fp8 model input issue (#724)
Wenhua Cheng committed 1 year ago
9113fa8
Fix hpu CI (#723)
Sun, Xuehao committed 1 year ago
173b092
Freeze torchvision version in CI (#720)
Sun, Xuehao committed 1 year ago
4b9fbed
Bump transformers in /test/test_cuda (#719)
dependabot[bot] committed 1 year ago
80a0917
Update step_by_step.md (#717)
Wenhua Cheng committed 1 year ago
e23cbf4
simply CpuInfo class (#715)
Wenhua Cheng committed 1 year ago
c731b65
61f0d82
fix critic bug of gguf tuning (#710)
Wenhua Cheng committed 1 year ago
d945afa