Xiuyu-Li q-diffusion issues

Xiuyu-Li / q-diffusion

[ICCV 2023] Q-Diffusion: Quantizing Diffusion Models.

https://xiuyuli.com/qdiffusion/

MIT License

290 stars 20 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Tensorrt employ the q-diffusion's recontruction method to quantize SD-XL?

#37 hanhanpp opened 2 months ago
0
Q-Diffusion support for Stable Diffusion XL (CPU)

#36 badhri-intel opened 2 months ago
0
How to do it in SDXL-turbo?

#35 ApolloRay opened 3 months ago
0
'functions' package error & how to extract FID

#34 parkjjoe opened 3 months ago
0
Can you please add a tutorial on quantification to the README？

#33 ningmenghongcha opened 3 months ago
0
BOPs measurement

#32 rocco-manz opened 3 months ago
0
Stable diffusion generated samples using quantized checkpoint looks strange

#31 colorjam opened 4 months ago
2
How to deploy in an environment without GPUs?

#30 loonglongdada closed 4 months ago
0
Compatibility with MacOSX?

#29 nachiket opened 6 months ago
0
如何获取量化参数

#28 miaott1234 opened 7 months ago
2
QDiffusion for Stable Diffusion

#27 stein-666 opened 7 months ago
0
Why quant QKMatMul as a block while it has no submodule?

#26 Sugar929 opened 8 months ago
6
What is the minimum number of datasets that meet the requirements of text-to-image calibration?

#25 hanhanpp opened 8 months ago
0
errors in executing in LSUN-bed w8a8 quantization

#24 Sugar929 opened 9 months ago
0
End-to-End Quantization for Speedup and Memory Savings: Inviting Contributions!

#23 Xiuyu-Li opened 9 months ago
2
Does q-diffusion work on SDXL?

#22 TruthSearcher closed 3 months ago
3
calibration datasets

#21 hanhanpp closed 8 months ago
0
Add calibration code

#20 Xiuyu-Li closed 9 months ago
0
Model sizes

#19 prinshul opened 9 months ago
1
Would you please provide the parameters for the LSQ and the block reconstruction? Thanks a lot

#18 yuzheyao22 closed 9 months ago
1
Open-source more code?

#17 lingffff closed 9 months ago
1
Question about the inference process

#16 JiaojiaoYe1994 opened 11 months ago
0
Why w4a8 quantization method have not accelerate the inferrence speed of Stable Diffusion models?

#15 felixslu closed 3 months ago
2
Why this quantization model need more than 24GB GPU memory which is larger than ideal 500M?

#14 felixslu opened 11 months ago
4
Error: No such file or directory: 'models/ldm/stable-diffusion-v1/model.ckpt'

#13 felixslu closed 11 months ago
1
About the quantized model

#12 shiyuetianqiang closed 3 months ago
0
load cifar_w8a8_ckpt.pth

#11 foreverlove944 closed 12 months ago
4
Code for model calibration

#10 Cheeun closed 9 months ago
5
模型大小

#9 ZhibinPeng closed 1 year ago
1
Loading quantized model checkpoint error

#8 arthursunbao closed 1 year ago
1
Extremely high VRAM usage

#7 easonoob closed 3 months ago
2
Images are broken in readme.md

#6 6174 closed 1 year ago
1
Calibration Process in the 'resume_cali_model' Function

#5 wangjialinEcopia closed 11 months ago
4
What is the main comparison of the paper

#4 sravanthOppo27 closed 1 year ago
1
the weight format of w4a8 is fp32?

#3 gongqiang closed 1 year ago
1
Activation quantization of Stable Diffusion.

#2 jianyuheng opened 1 year ago
0
Inference speed mechanism or Model size compression ?

#1 sravanthOppo27 closed 3 months ago
7