NetEase-FuXi EETQ issues

NetEase-FuXi / EETQ

Easy and Efficient Quantization for Transformers

Apache License 2.0

180 stars 14 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Bug: Import of shard_checkpoint from transformers fails

#34 BenjaminBossan opened 1 day ago
1
torch version issue

#33 sasakiyori opened 1 week ago
1
aarch64/ arm64 support

#32 khayamgondal opened 1 month ago
1
My system just updated CUDA to 12.6 and I can no longer compile EETQ. (Python 3.12)

#31 michael-newsrx opened 3 months ago
0
Unsupported Arch Assertion fail

#30 rahul3161 opened 3 months ago
2
安装之后报错

#29 LChuanwMz opened 3 months ago
1
EETQ-quantized TrOCR gives nonsense output

#28 donjuanpond closed 3 months ago
2
EETQ wheel not building

#27 donjuanpond opened 3 months ago
3
Does it support Whisper model?

#26 kadirnar closed 3 months ago
2
add qwen2

#25 ehartford opened 4 months ago
2
add Qwen2

#24 ehartford opened 4 months ago
6
ImportError: cannot import name 'EetqConfig' from 'transformers', despite using using 4.38.2 which satisfies >=4.27.0

#23 moruga123 closed 4 months ago
2
Repetition with Llama3-70b and EETQ

#22 mjsteele12 closed 2 months ago
2
Does it support Vision Transformers?

#21 PaulaDelgado-Santos closed 2 months ago
2
Create LICENSE

#20 dtlzhuangz closed 6 months ago
0
Support CPU quantization

#19 xgal opened 6 months ago
4
License

#18 AlpinDale closed 6 months ago
1
Qlora with eetq is quite slow

#17 hjh0119 opened 6 months ago
3
FIX: Use `matmul` instead of `mm` in `backward`

#16 younesbelkada closed 6 months ago
0
PEFT compatible GEMM

#15 dtlzhuangz closed 6 months ago
0
how to dequant a EETQ model?

#14 mxjmtxrm closed 6 months ago
4
Integration with Hugging Face transformers library

#13 younesbelkada closed 6 months ago
2
Supports H100

#12 mwbyeon closed 4 months ago
1
Modify code to support CUDA Graph

#11 jacob-crux closed 9 months ago
1
Quantization takes a very long time

#10 timohear opened 9 months ago
3
[docs] Update readme

#9 SidaZh closed 9 months ago
0
Add LoRAX to usage options in README.

#8 arnavgarg1 closed 9 months ago
4
rm dist

#7 dtlzhuangz closed 10 months ago
0
gemv optimization

#6 dtlzhuangz closed 10 months ago
0
Understanding EETQ and 8 bit quantization

#5 RonanKMcGovern closed 11 months ago
3
How to handle bfloat16?

#4 vgoklani closed 11 months ago
7
Why does EETQ take up all VRAM

#3 RonanKMcGovern closed 11 months ago
2
安装出错ERROR: Could not build wheels for EETQ, which is required to install pyproject.toml-based projects

#2 linshuijin closed 1 year ago
5
Question on outlier handling

#1 0xymoro closed 11 months ago
1