diff --git a/README.md b/README.md
index c8ab8a05fa7eadadcc1c997b6d561a8a23960c72..3952b694a0a52acbf1b8cea0e3585c199c74991b 100644
--- a/README.md
+++ b/README.md
@@ -1,13 +1,32 @@
-# Oneflow
-
-### Install OneFlow
-
-  #### System Requirements to Run OneFlow
+**OneFlow is a performance-centered and open-source platform for machine learning.**
+
+- [Install OneFlow](#install-oneflow)
+  - [System Requirements](#system-requirements)
+  - [Install with Pip Package](#install-with-pip-package)
+  - [Build from Source](#build-from-source)
+  - [Troubleshooting](#troubleshooting)
+  - [Advanced features](#advanced-features)
+- [Getting Started](#getting-started)
+- [Documentation](#documentation)
+    - [Usage & Design Docs](#usage--design-docs)
+    - [API Reference](#api-reference)
+- [Model Zoo and Benchmark](#model-zoo-and-benchmark)
+  - [CNNs(ResNet-50, VGG-16, Inception-V3, AlexNet)](#cnnsresnet-50-vgg-16-inception-v3-alexnet)
+  - [Wide&Deep](#widedeep)
+  - [BERT](#bert)
+- [Communication](#communication)
+- [Contributing](#contributing)
+- [The Team](#the-team)
+- [License](#license)
+
+## Install OneFlow
+
+  ### System Requirements
 
   - Python >= 3.5
   - Nvidia Linux x86_64 driver version >= 440.33
 
-  #### Install Pip package
+  ### Install with Pip Package
 
   - To install latest release of OneFlow with CUDA support:
 
@@ -36,7 +55,7 @@
 
   - Releases are built with G++/GCC 4.8.5, cuDNN 7 and MKL 2020.0-088.
 
-### Build OneFlow from Source
+### Build from Source
 
 1. #### System Requirements to Build OneFlow
 
@@ -94,8 +113,60 @@
 
 Please refer to [troubleshooting](docs/source/troubleshooting.md) for common issues you might encounter when compiling and running OneFlow.
 
-### Advanced Features
+### Advanced features
 
 - #### XRT
 
   You can check this [doc](oneflow/xrt/README.md) to obtain more details about how to use XLA and TensorRT with OneFlow.
+
+## Getting Started
+3 minutes to run MNIST.
+1. Clone the demo code from OneFlow documentation
+```
+git clone https://github.com/Oneflow-Inc/oneflow-documentation.git
+cd oneflow-documentation/cn/docs/code/quick_start/ 
+```
+2. Run it in Python
+```
+python mlp_mnist.py
+```
+
+3. Oneflow is running and you got the training loss
+```
+2.7290366
+0.81281316
+0.50629824
+0.35949975
+0.35245502
+...
+```
+More info on this demo, please refer to [doc on quick start](http://docs.oneflow.org/quick_start/quickstart_in_3_min.html).
+
+## Documentation
+#### Usage & Design Docs
+* [link](http://docs.oneflow.org/)
+#### API Reference
+* [link](https://oneflow-api.readthedocs.io/en/latest/)
+
+## Model Zoo and Benchmark
+* [link](https://github.com/Oneflow-Inc/OneFlow-Benchmark)
+### CNNs(ResNet-50, VGG-16, Inception-V3, AlexNet)
+* [CNNs](https://github.com/Oneflow-Inc/OneFlow-Benchmark/tree/of_develop_py3/Classification/cnns)
+
+### Wide&Deep
+* [OneFlow-WDL](https://github.com/Oneflow-Inc/OneFlow-Benchmark/tree/of_develop_py3/ClickThroughRate/WideDeepLearning)
+### BERT
+* [Bert](https://github.com/Oneflow-Inc/OneFlow-Benchmark/tree/of_develop_py3/LanguageModeling/BERT)
+
+## Communication
+* Github issues : any install, bug, feature issues.
+* [www.oneflow.org](http://www.oneflow.org) : brand related information.
+
+## Contributing
+*  [link](http://docs.oneflow.org/contribute/intro.html)
+
+## The Team
+OneFlow was originally developed by [OneFlow Inc](http://www.oneflow.org) and [Zhejiang Lab](http://www.zhejianglab.com/).
+
+## License
+[Apache License 2.0](LICENSE)