diff --git a/deploy/pdserving/clas_local_server.py b/deploy/pdserving/clas_local_server.py
index 4bceea771d33af452d05644e0fe517a79b475fc6..1d48adcfdb8c3543205f31b7d1345338c589ddcf 100644
--- a/deploy/pdserving/clas_local_server.py
+++ b/deploy/pdserving/clas_local_server.py
@@ -22,9 +22,9 @@ import time
 import re
 import base64
 from tools.infer.predict_cls import TextClassifier
-import tools.infer.utility as utility
+from params import read_params
 
-global_args = utility.parse_args()
+global_args = read_params()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
diff --git a/deploy/pdserving/clas_rpc_server.py b/deploy/pdserving/clas_rpc_server.py
index cc96199b5f7a8cddb64cf9a3c5b4e4e82e0ca95c..7fad61ee81361957568fbf8dd61e446106062084 100644
--- a/deploy/pdserving/clas_rpc_server.py
+++ b/deploy/pdserving/clas_rpc_server.py
@@ -22,9 +22,9 @@ import time
 import re
 import base64
 from tools.infer.predict_cls import TextClassifier
-import tools.infer.utility as utility
+from params import read_params
 
-global_args = utility.parse_args()
+global_args = read_params()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
diff --git a/deploy/pdserving/det_local_server.py b/deploy/pdserving/det_local_server.py
index d0e52cd047958cf9d4c28d8464c053ce406dbb68..2f7ee249502825e10358b7019e63fc45ce48d996 100644
--- a/deploy/pdserving/det_local_server.py
+++ b/deploy/pdserving/det_local_server.py
@@ -21,9 +21,9 @@ import time
 import re
 import base64
 from tools.infer.predict_det import TextDetector
-import tools.infer.utility as utility
+from params import read_params
 
-global_args = utility.parse_args()
+global_args = read_params()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
diff --git a/deploy/pdserving/det_rpc_server.py b/deploy/pdserving/det_rpc_server.py
index 6588a0d62826c714c9bc8ded45f9fe394bf68924..0d311cd974a48eb8cd03cda76da75ae3bf8ca0f8 100644
--- a/deploy/pdserving/det_rpc_server.py
+++ b/deploy/pdserving/det_rpc_server.py
@@ -21,9 +21,9 @@ import time
 import re
 import base64
 from tools.infer.predict_det import TextDetector
-import tools.infer.utility as utility
+from params import read_params
 
-global_args = utility.parse_args()
+global_args = read_params()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
diff --git a/deploy/pdserving/ocr_local_server.py b/deploy/pdserving/ocr_local_server.py
index b506790f2a3b7d6b4955bfc27640892f9104b24f..af9e72224d841d7451ef723aa207ebe0cb041ed8 100644
--- a/deploy/pdserving/ocr_local_server.py
+++ b/deploy/pdserving/ocr_local_server.py
@@ -24,12 +24,13 @@ import base64
 from clas_local_server import TextClassifierHelper
 from det_local_server import TextDetectorHelper
 from rec_local_server import TextRecognizerHelper
-import tools.infer.utility as utility
 from tools.infer.predict_system import TextSystem, sorted_boxes
 from paddle_serving_app.local_predict import Debugger
 import copy
+from params import read_params
+
+global_args = read_params()
 
-global_args = utility.parse_args()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
@@ -84,8 +85,7 @@ class TextSystemHelper(TextSystem):
 
 class OCRService(WebService):
     def init_rec(self):
-        args = utility.parse_args()
-        self.text_system = TextSystemHelper(args)
+        self.text_system = TextSystemHelper(global_args)
 
     def preprocess(self, feed=[], fetch=[]):
         # TODO: to handle batch rec images
diff --git a/deploy/pdserving/ocr_rpc_server.py b/deploy/pdserving/ocr_rpc_server.py
index 873e8792577da94ae99600100906211a956fc349..8229c141560423c8bff0dbec43877d6b1007d827 100644
--- a/deploy/pdserving/ocr_rpc_server.py
+++ b/deploy/pdserving/ocr_rpc_server.py
@@ -24,11 +24,11 @@ import base64
 from clas_rpc_server import TextClassifierHelper
 from det_rpc_server import TextDetectorHelper
 from rec_rpc_server import TextRecognizerHelper
-import tools.infer.utility as utility
 from tools.infer.predict_system import TextSystem, sorted_boxes
 import copy
+from params import read_params
 
-global_args = utility.parse_args()
+global_args = read_params()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
@@ -87,8 +87,7 @@ class TextSystemHelper(TextSystem):
 
 class OCRService(WebService):
     def init_rec(self):
-        args = utility.parse_args()
-        self.text_system = TextSystemHelper(args)
+        self.text_system = TextSystemHelper(global_args)
 
     def preprocess(self, feed=[], fetch=[]):
         # TODO: to handle batch rec images
diff --git a/deploy/pdserving/params.py b/deploy/pdserving/params.py
new file mode 100644
index 0000000000000000000000000000000000000000..6d0ecf988ff76db5784866639c8594428e326049
--- /dev/null
+++ b/deploy/pdserving/params.py
@@ -0,0 +1,50 @@
+# -*- coding:utf-8 -*-
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+class Config(object):
+    pass
+
+def read_params():
+    cfg = Config()
+    #use gpu
+    cfg.use_gpu = False
+    cfg.use_pdserving = True
+
+    #params for text detector
+    cfg.det_algorithm = "DB"
+    cfg.det_model_dir = "./det_mv_server/"
+    cfg.det_max_side_len = 960
+
+    #DB parmas
+    cfg.det_db_thresh =0.3
+    cfg.det_db_box_thresh =0.5
+    cfg.det_db_unclip_ratio =2.0
+
+    #EAST parmas
+    cfg.det_east_score_thresh = 0.8
+    cfg.det_east_cover_thresh = 0.1
+    cfg.det_east_nms_thresh = 0.2
+
+    #params for text recognizer
+    cfg.rec_algorithm = "CRNN"
+    cfg.rec_model_dir = "./ocr_rec_server/"
+
+    cfg.rec_image_shape = "3, 32, 320"
+    cfg.rec_char_type = 'ch'
+    cfg.rec_batch_num = 30
+    cfg.max_text_length = 25
+
+    cfg.rec_char_dict_path = "./ppocr_keys_v1.txt"
+    cfg.use_space_char = True
+
+    #params for text classifier
+    cfg.use_angle_cls = True
+    cfg.cls_model_dir = "./ocr_clas_server/"
+    cfg.cls_image_shape = "3, 48, 192"
+    cfg.label_list = ['0', '180']
+    cfg.cls_batch_num = 30
+    cfg.cls_thresh = 0.9
+
+    return cfg
diff --git a/deploy/pdserving/rec_local_server.py b/deploy/pdserving/rec_local_server.py
index 90f4a0e5f9af2ed7deedb9b5afd4fadbcedf8df4..b3ec74103bcbd37cfd4e8243c07731380597b3d5 100644
--- a/deploy/pdserving/rec_local_server.py
+++ b/deploy/pdserving/rec_local_server.py
@@ -22,9 +22,10 @@ import time
 import re
 import base64
 from tools.infer.predict_rec import TextRecognizer
-import tools.infer.utility as utility
+from params import read_params
+
+global_args = read_params()
 
-global_args = utility.parse_args()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
diff --git a/deploy/pdserving/rec_rpc_server.py b/deploy/pdserving/rec_rpc_server.py
index 38251b3e1957fe554ba667a8020d2eace85fd589..59282373a6733446293c6ac85192a77653faff67 100644
--- a/deploy/pdserving/rec_rpc_server.py
+++ b/deploy/pdserving/rec_rpc_server.py
@@ -22,9 +22,9 @@ import time
 import re
 import base64
 from tools.infer.predict_rec import TextRecognizer
-import tools.infer.utility as utility
+from params import read_params
 
-global_args = utility.parse_args()
+global_args = read_params()
 if global_args.use_gpu:
     from paddle_serving_server_gpu.web_service import WebService
 else:
diff --git a/doc/doc_ch/serving_inference.md b/doc/doc_ch/serving_inference.md
index 089424291525d631286725a8993582648d684385..2f77a2920278779dc126ed8d6ba891d63391e3f6 100644
--- a/doc/doc_ch/serving_inference.md
+++ b/doc/doc_ch/serving_inference.md
@@ -10,117 +10,100 @@
 
 ## 一、训练模型转Serving模型
 
-### 检测模型转Serving模型
+在前序文档 [基于Python预测引擎推理](./inference.md) 中，我们提供了如何把训练的checkpoint转换成Paddle模型。Paddle模型通常由一个文件夹构成，内含模型结构描述文件`model`和模型参数文件`params`。Serving模型由两个文件夹构成，用于存放客户端和服务端的配置。
 
-下载超轻量级中文检测模型：
+我们以`ch_rec_r34_vd_crnn`模型作为例子，下载链接在：
 
 ```
-wget -P ./ch_lite/ https://paddleocr.bj.bcebos.com/ch_models/ch_det_mv3_db.tar && tar xf ./ch_lite/ch_det_mv3_db.tar -C ./ch_lite/
+wget --no-check-certificate https://paddleocr.bj.bcebos.com/ch_models/ch_rec_r34_vd_crnn_infer.tar
+tar xf ch_rec_r34_vd_crnn_infer.tar
 ```
-
-上述模型是以MobileNetV3为backbone训练的DB算法，将训练好的模型转换成Serving模型只需要运行如下命令：
-
+因此我们按照Serving模型转换教程，运行下列python文件。
 ```
-# -c后面设置训练算法的yml配置文件
-# -o配置可选参数
-# Global.checkpoints参数设置待转换的训练模型地址，不用添加文件后缀.pdmodel，.pdopt或.pdparams。
-# Global.save_inference_dir参数设置转换的模型将保存的地址。
-
-python tools/export_serving_model.py -c configs/det/det_mv3_db.yml -o Global.checkpoints=./ch_lite/det_mv3_db/best_accuracy Global.save_inference_dir=./inference/det_db/
+python tools/inference_to_serving.py --model_dir ch_rec_r34_vd_crnn
 ```
-
-转Serving模型时，使用的配置文件和训练时使用的配置文件相同。另外，还需要设置配置文件中的`Global.checkpoints`、`Global.save_inference_dir`参数。 其中`Global.checkpoints`指向训练中保存的模型参数文件，`Global.save_inference_dir`是生成的inference模型要保存的目录。 转换成功后，在`save_inference_dir`目录下有两个文件：
+最终会在`serving_client_dir`和`serving_server_dir`生成客户端和服务端的模型配置。其中`serving_server_dir`和`serving_client_dir`的名字可以自定义。最终文件结构如下
 
 ```
-inference/det_db/
+/ch_rec_r34_vd_crnn/
 ├── serving_client_dir # 客户端配置文件夹
 └── serving_server_dir # 服务端配置文件夹
-
 ```
 
-### 识别模型转Serving模型
-
-下载超轻量中文识别模型：
-
-```
-wget -P ./ch_lite/ https://paddleocr.bj.bcebos.com/ch_models/ch_rec_mv3_crnn.tar && tar xf ./ch_lite/ch_rec_mv3_crnn.tar -C ./ch_lite/
-```
-
-识别模型转inference模型与检测的方式相同，如下：
+## 二、文本检测模型Serving推理
 
-```
-# -c后面设置训练算法的yml配置文件
-# -o配置可选参数
-# Global.checkpoints参数设置待转换的训练模型地址，不用添加文件后缀.pdmodel，.pdopt或.pdparams。
-# Global.save_inference_dir参数设置转换的模型将保存的地址。
+启动服务可以根据实际需求选择启动`标准版`或者`快速版`，两种方式的对比如下表：  
 
-python3 tools/export_serving_model.py -c configs/rec/rec_chinese_lite_train.yml -o Global.checkpoints=./ch_lite/rec_mv3_crnn/best_accuracy \
-        Global.save_inference_dir=./inference/rec_crnn/
-```
+|版本|特点|适用场景|
+|-|-|-|
+|标准版|稳定性高，分布式部署|适用于吞吐量大，需要跨机房部署的情况|
+|快速版|部署方便，预测速度快|适用于对预测速度要求高，迭代速度快的场景，Windows用户只能选择快速版|
 
-**注意：**如果您是在自己的数据集上训练的模型，并且调整了中文字符的字典文件，请注意修改配置文件中的`character_dict_path`是否是所需要的字典文件。
+接下来的命令中，我们会指定快速版和标准版的命令。需要说明的是，标准版只能用Linux平台，快速版可以支持Linux/Windows。
+文本检测模型推理，默认使用DB模型的配置参数，识别默认为CRNN。
 
-转换成功后，在目录下有两个文件：
+配置文件在`params.py`中，我们贴出配置部分，如果需要做改动，也在这个文件内部进行修改。
 
 ```
-/inference/rec_crnn/
-├── serving_client_dir # 客户端配置文件夹
-└── serving_server_dir # 服务端配置文件夹
-```
+def read_params():
+    cfg = Config()
+    #use gpu
+    cfg.use_gpu = False # 是否使用GPU
+    cfg.use_pdserving = True # 是否使用paddleserving，必须为True
 
-### 方向分类模型转Serving模型
+    #params for text detector
+    cfg.det_algorithm = "DB" # 检测算法， DB/EAST等
+    cfg.det_model_dir = "./det_mv_server/" # 检测算法模型路径
+    cfg.det_max_side_len = 960 
 
-下载方向分类模型：
+    #DB params
+    cfg.det_db_thresh =0.3
+    cfg.det_db_box_thresh =0.5
+    cfg.det_db_unclip_ratio =2.0
 
-```
-wget -P ./ch_lite/ https://paddleocr.bj.bcebos.com/20-09-22/cls/ch_ppocr_mobile-v1.1.cls_pre.tar && tar xf ./ch_lite/ch_ppocr_mobile-v1.1.cls_pre.tar -C ./ch_lite/
-```
+    #EAST params
+    cfg.det_east_score_thresh = 0.8
+    cfg.det_east_cover_thresh = 0.1
+    cfg.det_east_nms_thresh = 0.2
 
-方向分类模型转inference模型与检测的方式相同，如下：
+    #params for text recognizer
+    cfg.rec_algorithm = "CRNN" # 识别算法， CRNN/RARE等
+    cfg.rec_model_dir = "./ocr_rec_server/" # 识别算法模型路径
 
-```
-# -c后面设置训练算法的yml配置文件
-# -o配置可选参数
-# Global.checkpoints参数设置待转换的训练模型地址，不用添加文件后缀.pdmodel，.pdopt或.pdparams。
-# Global.save_inference_dir参数设置转换的模型将保存的地址。
+    cfg.rec_image_shape = "3, 32, 320"
+    cfg.rec_char_type = 'ch'
+    cfg.rec_batch_num = 30
+    cfg.max_text_length = 25
 
-python3 tools/export_serving_model.py -c configs/cls/cls_mv3.yml -o Global.checkpoints=./ch_lite/cls_model/best_accuracy \
-        Global.save_inference_dir=./inference/cls/
-```
+    cfg.rec_char_dict_path = "./ppocr_keys_v1.txt" # 识别算法字典文件
+    cfg.use_space_char = True
 
-转换成功后，在目录下有两个文件：
+    #params for text classifier
+    cfg.use_angle_cls = True # 是否启用分类算法
+    cfg.cls_model_dir = "./ocr_clas_server/" # 分类算法模型路径
+    cfg.cls_image_shape = "3, 48, 192"
+    cfg.label_list = ['0', '180']
+    cfg.cls_batch_num = 30
+    cfg.cls_thresh = 0.9
 
+    return cfg
 ```
-/inference/cls/
-├── serving_client_dir # 客户端配置文件夹
-└── serving_server_dir # 服务端配置文件夹
-```
-
-在接下来的教程中，我们将给出推理的demo模型下载链接。
+与本地预测不同的是，Serving预测需要一个客户端和一个服务端，因此接下来的教程都是两行代码。
 
+在正式执行服务端启动命令之前，先export PYTHONPATH到工程主目录下。
 ```
-wget --no-check-certificate https://paddleocr.bj.bcebos.com/deploy/pdserving/ocr_pdserving_suite.tar.gz
-tar zxf ocr_pdserving_suite.tar.gz
+export PYTHONPATH=$PWD:$PYTHONPATH
+cd deploy/pdserving
 ```
-
-
-
-## 二、文本检测模型Serving推理
-
-文本检测模型推理，默认使用DB模型的配置参数。当不使用DB模型时，在推理时，需要通过传入相应的参数进行算法适配，细节参考下文。
-
-与本地预测不同的是，Serving预测需要一个客户端和一个服务端，因此接下来的教程都是两行代码。所有的
-
 ### 1. 超轻量中文检测模型推理
 
 超轻量中文检测模型推理，可以执行如下命令启动服务端：
 
 ```
 #根据环境只需要启动其中一个就可以
-python det_rpc_server.py --use_pdserving True --det_model_dir det_mv_server #标准版，Linux用户
-python det_local_server.py --use_pdserving True --det_model_dir det_mv_server #快速版，Windows/Linux用户
+python det_rpc_server.py #标准版，Linux用户
+python det_local_server.py #快速版，Windows/Linux用户
 ```
-如果需要使用CPU版本，还需增加 `--use_gpu False`。
 
 客户端
 
@@ -129,23 +112,8 @@ python det_web_client.py
 ```
 
 
-
 Serving的推测和本地预测不同点在于，客户端发送请求到服务端，服务端需要检测到文字框之后返回框的坐标，此处没有后处理的图片，只能看到坐标值。
 
-### 2. DB文本检测模型推理
-
-首先将DB文本检测训练过程中保存的模型，转换成inference model。以基于Resnet50_vd骨干网络，在ICDAR2015英文数据集训练的模型为例（[模型下载地址](https://paddleocr.bj.bcebos.com/det_r50_vd_db.tar))，可以使用如下命令进行转换：
-
-```
-# -c后面设置训练算法的yml配置文件
-# Global.checkpoints参数设置待转换的训练模型地址，不用添加文件后缀.pdmodel，.pdopt或.pdparams。
-# Global.save_inference_dir参数设置转换的模型将保存的地址。
-
-python3 tools/export_serving_model.py -c configs/det/det_r50_vd_db.yml -o Global.checkpoints="./models/det_r50_vd_db/best_accuracy" Global.save_inference_dir="./inference/det_db"
-```
-
-经过转换之后，会在`./inference/det_db` 目录下出现`serving_server_dir`和`serving_client_dir`，然后指定`det_model_dir` 。
-
 ## 三、文本识别模型Serving推理
 
 下面将介绍超轻量中文识别模型推理、基于CTC损失的识别模型推理和基于Attention损失的识别模型推理。对于中文文本识别，建议优先选择基于CTC损失的识别模型，实践中也发现基于Attention损失的效果不如基于CTC损失的识别模型。此外，如果训练时修改了文本的字典，请参考下面的自定义文本识别字典的推理。
@@ -153,11 +121,11 @@ python3 tools/export_serving_model.py -c configs/det/det_r50_vd_db.yml -o Global
 ### 1. 超轻量中文识别模型推理
 
 超轻量中文识别模型推理，可以执行如下命令启动服务端：
-
+需要注意params.py中的`--use_gpu`的值
 ```
 #根据环境只需要启动其中一个就可以
-python rec_rpc_server.py --use_pdserving True --rec_model_dir ocr_rec_server #标准版，Linux用户
-python rec_local_server.py --use_pdserving True --rec_model_dir ocr_rec_server #快速版，Windows/Linux用户
+python rec_rpc_server.py #标准版，Linux用户
+python rec_local_server.py #快速版，Windows/Linux用户
 ```
 如果需要使用CPU版本，还需增加 `--use_gpu False`。
 
@@ -186,13 +154,12 @@ python rec_web_client.py
 ### 1. 方向分类模型推理
 
 方向分类模型推理， 可以执行如下命令启动服务端：
-
+需要注意params.py中的`--use_gpu`的值
 ```
 #根据环境只需要启动其中一个就可以
-python clas_rpc_server.py --use_pdserving True --cls_model_dir ocr_clas_server #标准版，Linux用户
-python clas_local_server.py --use_pdserving True --cls_model_dir ocr_clas_server #快速版，Windows/Linux用户
+python clas_rpc_server.py #标准版，Linux用户
+python clas_local_server.py #快速版，Windows/Linux用户
 ```
-如果需要使用CPU版本，还需增加 `--use_gpu False`。
 
 客户端
 
@@ -216,20 +183,20 @@ python rec_web_client.py
 在执行预测时，需要通过参数`image_dir`指定单张图像或者图像集合的路径、参数`det_model_dir`,`cls_model_dir`和`rec_model_dir`分别指定检测，方向分类和识别的inference模型路径。参数`use_angle_cls`用于控制是否启用方向分类模型。与本地预测不同的是，为了减少网络传输耗时，可视化识别结果目前不做处理，用户收到的是推理得到的文字字段。
 
 执行如下命令启动服务端：
-
+需要注意params.py中的`--use_gpu`的值
 ```
 #标准版，Linux用户
 #GPU用户
 python -m paddle_serving_server_gpu.serve --model det_mv_server --port 9293 --gpu_id 0
 python -m paddle_serving_server_gpu.serve --model ocr_cls_server --port 9294 --gpu_id 0
-python ocr_rpc_server.py --use_pdserving True --use_gpu True --rec_model_dir ocr_rec_server
+python ocr_rpc_server.py 
 #CPU用户
 python -m paddle_serving_server.serve --model det_mv_server --port 9293
 python -m paddle_serving_server.serve --model ocr_cls_server --port 9294
-python ocr_rpc_server.py --use_pdserving True --use_gpu False --rec_model_dir ocr_rec_server
+python ocr_rpc_server.py
 
 #快速版，Windows/Linux用户
-python ocr_local_server.py --use_gpu False --use_pdserving True --rec_model_dir ocr_rec_server/ --det_model_dir det_mv_server/ --cls_model_dir ocr_clas_server/ --rec_char_dict_path ppocr_keys_v1.txt  --use_angle_cls True
+python ocr_local_server.py 
 ```
 
 客户端
diff --git a/ppocr/data/det/db_process.py b/ppocr/data/det/db_process.py
index 1d3dafe54c6c4d30af8fd0ba7e7c1423b02a600a..bbd6174c760ab50a5f6fd301e594514b4431ba44 100644
--- a/ppocr/data/det/db_process.py
+++ b/ppocr/data/det/db_process.py
@@ -21,12 +21,10 @@ from ppocr.utils.utility import initial_logger, check_and_read_gif
 logger = initial_logger()
 
 import tools.infer.utility as utility
-args = utility.parse_args()
-if args.use_pdserving is False:
-    from .data_augment import AugmentData
-    from .random_crop_data import RandomCropData
-    from .make_shrink_map import MakeShrinkMap
-    from .make_border_map import MakeBorderMap
+from .data_augment import AugmentData
+from .random_crop_data import RandomCropData
+from .make_shrink_map import MakeShrinkMap
+from .make_border_map import MakeBorderMap
 
 
 class DBProcessTrain(object):
diff --git a/tools/export_serving_model.py b/tools/export_serving_model.py
deleted file mode 100644
index b43fac67999de1e39e6439300afa8867880b55c2..0000000000000000000000000000000000000000
--- a/tools/export_serving_model.py
+++ /dev/null
@@ -1,78 +0,0 @@
-# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
-#
-# Licensed under the Apache License, Version 2.0 (the "License");
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-from __future__ import absolute_import
-from __future__ import division
-from __future__ import print_function
-
-import os
-import sys
-__dir__ = os.path.dirname(os.path.abspath(__file__))
-sys.path.append(__dir__)
-sys.path.append(os.path.abspath(os.path.join(__dir__, '..')))
-
-
-def set_paddle_flags(**kwargs):
-    for key, value in kwargs.items():
-        if os.environ.get(key, None) is None:
-            os.environ[key] = str(value)
-
-
-# NOTE(paddle-dev): All of these flags should be
-# set before `import paddle`. Otherwise, it would
-# not take any effect.
-set_paddle_flags(
-    FLAGS_eager_delete_tensor_gb=0,  # enable GC to save memory
-)
-
-import program
-from paddle import fluid
-from ppocr.utils.utility import initial_logger
-logger = initial_logger()
-from ppocr.utils.save_load import init_model
-from paddle_serving_client.io import save_model
-
-
-def main():
-    startup_prog, eval_program, place, config, _ = program.preprocess()
-
-    feeded_var_names, target_vars, fetches_var_name = program.build_export(
-        config, eval_program, startup_prog)
-    eval_program = eval_program.clone(for_test=True)
-    exe = fluid.Executor(place)
-    exe.run(startup_prog)
-
-    init_model(config, eval_program, exe)
-
-    save_inference_dir = config['Global']['save_inference_dir']
-    if not os.path.exists(save_inference_dir):
-        os.makedirs(save_inference_dir)
-    serving_client_dir = "{}/serving_client_dir".format(save_inference_dir)
-    serving_server_dir = "{}/serving_server_dir".format(save_inference_dir)
-
-    feed_dict = {
-        x: eval_program.global_block().var(x)
-        for x in feeded_var_names
-    }
-    fetch_dict = {x.name: x for x in target_vars}
-    save_model(serving_server_dir, serving_client_dir, feed_dict, fetch_dict,
-               eval_program)
-    print(
-        "paddle serving model saved in {}/serving_server_dir and {}/serving_client_dir".
-        format(save_inference_dir, save_inference_dir))
-    print("save success, output_name_list:", fetches_var_name)
-
-
-if __name__ == '__main__':
-    main()
diff --git a/tools/infer/predict_cls.py b/tools/infer/predict_cls.py
index d5c7830b9e9d6661709642eefacf1819d3f137c9..dc4028d0ddcafa8e69cc47c783efe067fda51a0b 100755
--- a/tools/infer/predict_cls.py
+++ b/tools/infer/predict_cls.py
@@ -36,10 +36,10 @@ class TextClassifier(object):
         if args.use_pdserving is False:
             self.predictor, self.input_tensor, self.output_tensors = \
                 utility.create_predictor(args, mode="cls")
+            self.use_zero_copy_run = args.use_zero_copy_run
         self.cls_image_shape = [int(v) for v in args.cls_image_shape.split(",")]
         self.cls_batch_num = args.rec_batch_num
         self.label_list = args.label_list
-        self.use_zero_copy_run = args.use_zero_copy_run
         self.cls_thresh = args.cls_thresh
 
     def resize_norm_img(self, img):
diff --git a/tools/infer/predict_det.py b/tools/infer/predict_det.py
index 4dd9fdf78efd1e3e991b3c8713b4a747ffce2f74..60d67c963a262af8a15cd9cae8b8540090bbedac 100755
--- a/tools/infer/predict_det.py
+++ b/tools/infer/predict_det.py
@@ -42,7 +42,6 @@ class TextDetector(object):
     def __init__(self, args):
         max_side_len = args.det_max_side_len
         self.det_algorithm = args.det_algorithm
-        self.use_zero_copy_run = args.use_zero_copy_run
         preprocess_params = {'max_side_len': max_side_len}
         postprocess_params = {}
         if self.det_algorithm == "DB":
@@ -76,6 +75,7 @@ class TextDetector(object):
             logger.info("unknown det_algorithm:{}".format(self.det_algorithm))
             sys.exit(0)
         if args.use_pdserving is False:
+            self.use_zero_copy_run = args.use_zero_copy_run
             self.predictor, self.input_tensor, self.output_tensors =\
                 utility.create_predictor(args, mode="det")
 
diff --git a/tools/infer/predict_rec.py b/tools/infer/predict_rec.py
index e9fa52ac646c8194da9dc2c4fdc388011860d991..1989f524c21d519f879290bcc72bc1b282a93772 100755
--- a/tools/infer/predict_rec.py
+++ b/tools/infer/predict_rec.py
@@ -37,12 +37,12 @@ class TextRecognizer(object):
         if args.use_pdserving is False:
             self.predictor, self.input_tensor, self.output_tensors =\
                 utility.create_predictor(args, mode="rec")
+            self.use_zero_copy_run = args.use_zero_copy_run
         self.rec_image_shape = [int(v) for v in args.rec_image_shape.split(",")]
         self.character_type = args.rec_char_type
         self.rec_batch_num = args.rec_batch_num
         self.rec_algorithm = args.rec_algorithm
         self.text_len = args.max_text_length
-        self.use_zero_copy_run = args.use_zero_copy_run
         char_ops_params = {
             "character_type": args.rec_char_type,
             "character_dict_path": args.rec_char_dict_path,
diff --git a/tools/infer/utility.py b/tools/infer/utility.py
index 7dbeeec3751d0f86d754648fb19ede7490e2d862..1eb57d34490edd3557dbe09a541b0a173ac7bbdf 100755
--- a/tools/infer/utility.py
+++ b/tools/infer/utility.py
@@ -37,7 +37,6 @@ def parse_args():
     parser.add_argument("--ir_optim", type=str2bool, default=True)
     parser.add_argument("--use_tensorrt", type=str2bool, default=False)
     parser.add_argument("--gpu_mem", type=int, default=8000)
-    parser.add_argument("--use_pdserving", type=str2bool, default=False)
 
     # params for text detector
     parser.add_argument("--image_dir", type=str)
diff --git a/tools/inference_to_serving.py b/tools/inference_to_serving.py
new file mode 100644
index 0000000000000000000000000000000000000000..77af59cc70a2a6e7610579d798f2722c3c63f847
--- /dev/null
+++ b/tools/inference_to_serving.py
@@ -0,0 +1,29 @@
+# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import argparse
+from paddle_serving_client.io import inference_model_to_serving
+def parse_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--model_dir", type=str)
+    parser.add_argument("--server_dir", type=str, default="serving_server_dir")
+    parser.add_argument("--client_dir", type=str, default="serving_client_dir")
+    return parser.parse_args()
+
+args = parse_args()
+inference_model_dir = args.model_dir
+serving_client_dir = args.server_dir
+serving_server_dir = args.client_dir
+feed_var_names, fetch_var_names = inference_model_to_serving(
+        inference_model_dir, serving_client_dir, serving_server_dir, model_filename="model", params_filename="params")