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

update readme

wenhuach21bb0e4a5c311b
diff --git a/README.md b/README.md
index 80a5f1760..610a3a73b 100644
--- a/README.md
+++ b/README.md
@@ -71,9 +71,7 @@ We are working on exporting the quantized model to HF format
Cpu kernel will be supported soon
## Validated Models
-
-![](./figs/W4G-1.png)
-
+For a fair comparison, we utilized 512 samples from Pile-10k for all methods during calibration. Due to memory constraints, we maintained the original sequence length of 512 for AWQ, while for GPTQ and our approach, a sequence length of 2048 is used. The notation GPTQ* indicates that we adjusted the random seed or data preprocessing to address issues related to the in-positive Hessian matrix or other issues.
![](./figs/W4G128.png)
![](./figs/W3G128.png)
![](./figs/W2G128.png)