diff --git a/configs/ppyolo/README.md b/configs/ppyolo/README.md
index 9f529a1d617884b228f7917baf453991954e5b17..6ec39a5141ca6fb5191d66a6d727352fa165a0a6 100644
--- a/configs/ppyolo/README.md
+++ b/configs/ppyolo/README.md
@@ -67,7 +67,7 @@ PP-YOLO improved performance and speed of YOLOv3 with following methods:
 ### PP-YOLO for mobile
 
 |            Model             | GPU number | images/GPU | Model Size | input shape | Box AP<sup>val</sup> | Kirin 990(FPS) | download | config  |
-|:-------------  -------------:|:----------:|:----------:| :--------: | :----------:| :------------------: | :------------: | :------: | :-----: |
+|:----------------------------:|:----------:|:----------:| :--------: | :----------:| :------------------: | :------------: | :------: | :-----: |
 | PP-YOLO_MobileNetV3_large    |     4      |      32    |    18MB    |     320     |         22.0         |      14.1      | [model](https://paddlemodels.bj.bcebos.com/object_detection/ppyolo_mobilenet_v3_large.pdparams) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/master/configs/ppyolo/ppyolo_mobilenet_v3_large.yml)                   |
 | PP-YOLO_MobileNetV3_small    |     4      |      32    |    11MB    |     320     |         16.8         |      21.5      | [model](https://paddlemodels.bj.bcebos.com/object_detection/ppyolo_mobilenet_v3_small.pdparams) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/master/configs/ppyolo/ppyolo_mobilenet_v3_small.yml)                   |
 
diff --git a/configs/ppyolo/README_cn.md b/configs/ppyolo/README_cn.md
index 2322d34fc4c974f0fbb7f03b397bb0b61ec7a989..79c21e91c496d35d54b980bcb47020d018d99e1b 100644
--- a/configs/ppyolo/README_cn.md
+++ b/configs/ppyolo/README_cn.md
@@ -65,10 +65,10 @@ PP-YOLO从如下方面优化和提升YOLOv3模型的精度和速度：
 - PP-YOLO模型推理速度测试采用单卡V100，batch size=1进行测试，使用CUDA 10.2, CUDNN 7.5.1，TensorRT推理速度测试使用TensorRT 5.1.2.2。
 - YOLOv4(AlexyAB)行`模型下载`和`配置文件`为PaddleDetection复现的YOLOv4模型，目前评估精度已对齐，支持finetune，训练精度对齐中，可参见[PaddleDetection YOLOv4 模型](../yolov4/README.md)
 
-### PP-YOLO 移动端模型
+### PP-YOLO 轻量级模型
 
 |          模型                | GPU个数 | 每GPU图片个数 |  模型体积  | 输入尺寸 | Box AP<sup>val</sup> | Kirin 990 (FPS) | 模型下载 | 配置文件 |
-|:----------------------------:|:-------:|:-------------:|:----------:| :-------:| :------------------: | :-------------: |------------: | :---------------------: | :------: | :------: |
+|:----------------------------:|:-------:|:-------------:|:----------:| :-------:| :------------------: | :-------------: |------------: | :---------------------: |
 | PP-YOLO_MobileNetV3_large    |    4    |      32       |    18MB    |   320    |         22.0         |       14.1      | [下载链接](https://paddlemodels.bj.bcebos.com/object_detection/ppyolo_mobilenet_v3_large.pdparams) | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/master/configs/ppyolo/ppyolo_mobilenet_v3_large.yml)                   |
 | PP-YOLO_MobileNetV3_small    |    4    |      32       |    11MB    |   320    |         16.8         |       21.5      | [下载链接](https://paddlemodels.bj.bcebos.com/object_detection/ppyolo_mobilenet_v3_small.pdparams) | [配置文件](https://github.com/PaddlePaddle/PaddleDetection/tree/master/configs/ppyolo/ppyolo_mobilenet_v3_small.yml)                   |
 
diff --git a/configs/ppyolo/ppyolo_mobilenet_v3_large.yml b/configs/ppyolo/ppyolo_mobilenet_v3_large.yml
index cb1d5cbc272dd78c5a134b7b8966855805c9fcb7..cc6dd34773e23ec76bb9eb36d0852b5f318b2de7 100755
--- a/configs/ppyolo/ppyolo_mobilenet_v3_large.yml
+++ b/configs/ppyolo/ppyolo_mobilenet_v3_large.yml
@@ -130,9 +130,8 @@ TrainReader:
   shuffle: true
   mixup_epoch: 500
   drop_last: true
-  worker_num: 16
-  bufsize: 16
-  memsize: 8G
+  worker_num: 8
+  bufsize: 4
   use_process: true
 
 EvalReader:
diff --git a/configs/ppyolo/ppyolo_mobilenet_v3_small.yml b/configs/ppyolo/ppyolo_mobilenet_v3_small.yml
index 5f8cf034efa675b5715d652b2e32a5d3d3ecca6b..3e189cc58dec1790adce33bfdf372bec5b0cc895 100755
--- a/configs/ppyolo/ppyolo_mobilenet_v3_small.yml
+++ b/configs/ppyolo/ppyolo_mobilenet_v3_small.yml
@@ -130,9 +130,8 @@ TrainReader:
   shuffle: true
   mixup_epoch: 500
   drop_last: true
-  worker_num: 16
-  bufsize: 16
-  memsize: 8G
+  worker_num: 8
+  bufsize: 4
   use_process: true
 
 EvalReader: