diff --git a/docs/mindspore/source_zh_cn/model_infer/ms_infer/model_export.md b/docs/mindspore/source_zh_cn/model_infer/ms_infer/model_export.md
index c10c0e112a086b057c8aa9a144e91edeaec33227..b500148d51c751d56bba6db859d38f7f1859ede7 100644
--- a/docs/mindspore/source_zh_cn/model_infer/ms_infer/model_export.md
+++ b/docs/mindspore/source_zh_cn/model_infer/ms_infer/model_export.md
@@ -1,3 +1,22 @@
 # 模型导出
 
 [![查看源文件](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_source.svg)](https://gitee.com/mindspore/docs/blob/master/docs/mindspore/source_zh_cn/model_infer/ms_infer/model_export.md)
+
+MindSpore提供了云侧（训练）和端侧（推理）统一的中间表示（Intermediate Representation, IR）。可使用export接口直接将模型保存为MindIR（当前仅支持严格图模式）。
+
+```python
+import mindspore as ms
+import numpy as np
+from mindspore import Tensor
+
+# Define the network structure of LeNet5. Refer to
+# https://gitee.com/mindspore/docs/blob/r2.3.1/docs/mindspore/code/lenet.py
+net = LeNet5()
+input_tensor = Tensor(np.ones([1, 1, 32, 32]).astype(np.float32))
+ms.export(net, input_tensor, file_name='lenet', file_format='MINDIR')
+
+```
+
+详细接口参考链接：[mindspore.export](https://www.mindspore.cn/docs/zh-CN/r2.3.1/api_python/mindspore/mindspore.export.html?highlight=export#mindspore.export)。
+
+模型导出结果提供给MindSpore Lite使用，使用方式可参考[使用Python接口执行云侧推理](https://www.mindspore.cn/lite/docs/zh-CN/r2.3.1/use/cloud_infer/runtime_python.html)。
diff --git a/docs/mindspore/source_zh_cn/model_infer/ms_infer/profiling.md b/docs/mindspore/source_zh_cn/model_infer/ms_infer/profiling.md
index a613dc7935d38e07cee423e4af6d9ae7adc52f2b..4c4cc850f1f9573e423662a00a88539d9f9951a9 100644
--- a/docs/mindspore/source_zh_cn/model_infer/ms_infer/profiling.md
+++ b/docs/mindspore/source_zh_cn/model_infer/ms_infer/profiling.md
@@ -1,3 +1,32 @@
 # Profiling
 
 [![查看源文件](https://mindspore-website.obs.cn-north-4.myhuaweicloud.com/website-images/master/resource/_static/logo_source.svg)](https://gitee.com/mindspore/docs/blob/master/docs/mindspore/source_zh_cn/model_infer/ms_infer/profiling.md)
+
+MindSpore中提供了profiler接口，可以对神经网络的性能进行采集。目前支持AICORE算子、AICPU算子、HostCPU算子、内存、设备通信、集群等数据的分析。
+
+样例：
+
+```python
+import numpy as np
+import mindspore
+from mindspore import Tensor
+from mindspore.train import Model
+from mindspore import Profiler
+
+input_data = Tensor(np.random.randint(0, 255, [1, 1, 32, 32]), mindspore.float32)
+# Define the network structure of LeNet5. Refer to
+# https://gitee.com/mindspore/docs/blob/r2.3.1/docs/mindspore/code/lenet.py
+# Init Profiler
+# Note that the Profiler should be initialized before model.predict
+profiler = Profiler()
+model = Model(LeNet5())
+result = model.predict(input_data)
+
+# Profiler end
+profiler.analyse()
+
+```
+
+推理方面性能调试方式与训练基本一致，收集到性能数据后，可参考：[性能调试（Ascend）](https://www.mindspore.cn/mindinsight/docs/zh-CN/master/performance_profiling_ascend.html)进行性能分析。推理上重点关注算子性能分析、计算量性能分析、Timeline分析等。
+
+详细接口参考：[mindspore.Profiler](https://www.mindspore.cn/docs/zh-CN/r2.3.1/api_python/mindspore/mindspore.Profiler.html?highlight=profiler#mindspore.Profiler)。