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

fix gguf fp8 input model and vram issue (#844)
Wenhua Cheng committed 1 year ago
4562cf4
[High Risk]Refine inference code (#840)
Wenhua Cheng committed 1 year ago
162424d
fix log error (#843)
Heng Guo committed 1 year ago
4b41826
77c608f
dd7811e
fix hpu detect issue (#823)
Xin He committed 1 year ago
6e8732c
hpu build with auto_round package name (#838)
chen, suyue committed 1 year ago
a280485
MXFP4 and MXFP8 loading support (#832)
Yi Liu committed 1 year ago
f44d1f7
fix format checking bug (#836)
Weiwei committed 1 year ago
81c1f52
bump version to 0.8.0.dev (#830)
Sun, Xuehao committed 1 year ago
95973a4
rm triton from requirements.txt (#824)
Wenhua Cheng committed 1 year ago
c898d13
d74f4bc
Fix llm-compressor export (#820)
Tang Kaihui committed 1 year ago
40aed06
Refine logger and add envs (#817)
Yi Liu committed 1 year ago
c5302e1
fix nsamples in get_dataloader (#804)
Wenhua Cheng committed 1 year ago
00fff81
Bump release version to v0.7.0 (#814)
Sun, Xuehao committed 1 year ago
30d1061
fix gptqmodel inference issue (#813)
Wenhua Cheng committed 1 year ago
c4e7fbe
fix bug of q_layer_inputs (#811)
Heng Guo committed 1 year ago
e1ae6e9
remove torch compile (#807)
Wenhua Cheng committed 1 year ago
1452d95
886d657
fix mxfp exporting (#806)
Weiwei committed 1 year ago
520c78f
improve the robustness of scheme (#803)
Wenhua Cheng committed 1 year ago
9534d2a
enable model python files saving (#802)
Weiwei committed 1 year ago
da089e9
fix several cuda ut bug (#797)
Heng Guo committed 1 year ago
ab55a97
fix typo in readme (#799)
Wenhua Cheng committed 1 year ago
e39f25e