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

924a468
Update get_block_names func (#1047)
Wang, Mengni committed 1 year ago
548acbb
247a9a7
fix accuracy regression (#1041)
Wenhua Cheng committed 1 year ago
6a8ac7e
Bump into v0.9.0 (#1024)
Sun, Xuehao committed 1 year ago
8d8a1cd
fix incorrect model name (#1035)
Wenhua Cheng committed 1 year ago
509ac2c
optimize vram for gguf and add momentum (#1031)
Wenhua Cheng committed 1 year ago
58b3d90
update gguf alg ext (#1026)
Heng Guo committed 1 year ago
81caded
Reduce mem usage of GPT-OSS (#1013)
Yi Liu committed 1 year ago
6d3d87d
Fix the missing temporary name (#1029)
Yi Liu committed 1 year ago
c4d030d
disable itrex format first (#998)
Weiwei committed 1 year ago
395602b
fix cuda ut (#1020)
Heng Guo committed 1 year ago
797f476
255322f
6d03ec8
fix multi cuda ut bug (#1014)
Heng Guo committed 1 year ago
d00cc85
5c2f1bd
fix bug (#1011)
Wenhua Cheng committed 1 year ago
e3f9610
Minor refactor for LLMC (#993)
Yi Liu committed 1 year ago
6af724a
add device list to clear memory (#1009)
Wenhua Cheng committed 1 year ago
1d02da1
Fix param missing bug (#1008)
Wang, Mengni committed 1 year ago
7f9fa0e
fix cuda CI (#1010)
Tang Kaihui committed 1 year ago
b7ef006
Fix non auto device map (#1005)
Weiwei committed 1 year ago
cd54e70
ead6f29
Fix diffusion multi-device ut issue (#1002)
Wang, Mengni committed 1 year ago
f6745fd
daeb3bb
Refine exllamav2 ut (#1001)
Weiwei committed 1 year ago
758b239