diff --git a/paddle/contrib/inference/high_level_api.md b/paddle/contrib/inference/high_level_api.md
new file mode 100644
index 0000000000000000000000000000000000000000..563b696143de9cbf67db38048bbd2f7c11b3a66e
--- /dev/null
+++ b/paddle/contrib/inference/high_level_api.md
@@ -0,0 +1,59 @@
+# Inference High-level APIs
+This document describes the high-level inference APIs one can use to easily deploy a Paddle model for an application.
+
+The APIs are described in `paddle_inference_api.h`, just one header file, and two libaries `libpaddle_fluid.so` and `libpaddle_fluid_api.so` are needed.
+
+## PaddleTensor
+We provide the `PaddleTensor` data structure is to give a general tensor interface.
+
+The definition is 
+
+```c++
+struct PaddleTensor {
+  std::string name;  // variable name.
+  std::vector<int> shape;
+  PaddleBuf data;  // blob of data.
+  PaddleDType dtype;
+};
+```
+
+The data is stored in a continuous memory `PaddleBuf`, and tensor's data type is specified by a `PaddleDType`. 
+The `name` field is used to specify the name of input variable, 
+that is important when there are multiple inputs and need to distiuish which variable to set.
+
+## engine
+The inference APIs has two different underlying implementation, currently there are two valid engines:
+
+- the native engine, which is consists of the native operators and framework,
+- the Anakin engine, which is a Anakin library embeded.
+
+The native engine takes a native Paddle model as input, and supports any model that trained by Paddle, 
+but the Anakin engine can only take the Anakin model as input(user need to manully transform the format first) and currently not all Paddle models are supported.
+
+```c++
+enum class PaddleEngineKind {
+  kNative = 0,  // Use the native Fluid facility.
+  kAnakin,      // Use Anakin for inference.
+};
+```
+
+## PaddlePredictor and how to create one
+The main interface is `PaddlePredictor`, there are following methods 
+
+- `bool Run(const std::vector<PaddleTensor>& inputs, std::vector<PaddleTensor>* output_data)`
+  - take inputs and output `output_data`
+- `Clone` to clone a predictor from an existing one, with model parameter shared.
+
+There is a factory method to help create a predictor, and the user takes the ownership of this object.
+
+```c++
+template <typename ConfigT, PaddleEngineKind engine = PaddleEngineKind::kNative>
+std::unique_ptr<PaddlePredictor> CreatePaddlePredictor(const ConfigT& config);
+```
+
+By specifying the engine kind and config, one can get an specific implementation.
+
+## Reference
+
+- [paddle_inference_api.h](./paddle_inference_api.h)
+- [demos](./demo)
diff --git a/paddle/contrib/inference/high_level_api_cn.md b/paddle/contrib/inference/high_level_api_cn.md
new file mode 100644
index 0000000000000000000000000000000000000000..a57f015a4e44d43ee4e475cf606faa6f05e095fa
--- /dev/null
+++ b/paddle/contrib/inference/high_level_api_cn.md
@@ -0,0 +1,87 @@
+# Paddle 预测 API
+
+为了更简单方便的预测部署，Fluid 提供了一套高层 API 用来隐藏底层不同的优化实现。
+
+预测库包含:
+
+- 头文件 `paddle_inference_api.h` 定义了所有的接口
+- 库文件`libpaddle_fluid.so` 或 `libpaddle_fluid.a`
+- 库文件 `libpaddle_inference_api.so` 或 `libpaddle_inference_api.a`
+
+下面是详细的一些 API 概念介绍
+
+## PaddleTensor
+
+PaddleTensor 定义了预测最基本的输入输出的数据格式，其定义是
+
+```c++
+struct PaddleTensor {
+  std::string name;  // variable name.
+  std::vector<int> shape;
+  PaddleBuf data;  // blob of data.
+  PaddleDType dtype;
+};
+```
+
+- `name` 用于指定输入数据对应的 模型中variable 的名字 （暂时没有用，但会在后续支持任意 target 时启用）
+- `shape` 表示一个 Tensor 的 shape
+- `data`  数据以连续内存的方式存储在`PaddleBuf` 中，`PaddleBuf` 可以接收外面的数据或者独立`malloc`内存，详细可以参考头文件中相关定义。
+- `dtype` 表示 Tensor 的数据类型
+
+## engine
+
+高层 API 底层有多种优化实现，我们称之为 engine，目前有三种 engine
+
+- 原生 engine，由 paddle 原生的 forward operator 组成，可以天然支持所有paddle 训练出的模型，
+- Anakin engine，封装了 [Anakin](https://github.com/PaddlePaddle/Anakin) ，在某些模型上性能不错，但只能接受自带模型格式，无法支持所有 paddle 模型，
+- TensorRT mixed engine，用子图的方式支持了 [TensorRT](https://developer.nvidia.com/tensorrt) ，支持所有paddle 模型，并自动切割部分计算子图到 TensorRT 上加速（WIP）
+
+其实现为
+
+```c++
+enum class PaddleEngineKind {
+  kNative = 0,       // Use the native Fluid facility.
+  kAnakin,           // Use Anakin for inference.
+  kAutoMixedTensorRT // Automatically mixing TensorRT with the Fluid ops.
+};
+```
+
+## 预测部署过程
+
+总体上分为以下步骤
+
+1. 用合适的配置创建 `PaddlePredictor`
+2. 创建输入用的 `PaddleTensor`，传入到 `PaddlePredictor` 中
+3. 获取输出的 `PaddleTensor` ，将结果取出
+
+下面完整演示一个简单的模型，部分细节代码隐去
+
+```c++
+#include "paddle_inference_api.h"
+
+// 创建一个 config，并修改相关设置
+paddle::NativeConfig config;
+config.model_dir = "xxx";
+config.use_gpu = false;
+// 创建一个原生的 PaddlePredictor
+auto predictor =
+      paddle::CreatePaddlePredictor<NativeConfig, PaddleEngineKind::kNative>(config);
+// 创建输入 tensor
+int64_t data[4] = {1, 2, 3, 4};
+paddle::PaddleTensor tensor{.name = "",
+                            .shape = std::vector<int>({4, 1}),
+                            .data = PaddleBuf(data, sizeof(data)),
+                            .dtype = PaddleDType::INT64};
+// 创建输出 tensor，输出 tensor 的内存可以复用
+std::vector<paddle::PaddleTensor> outputs;
+// 执行预测
+CHECK(predictor->Run(slots, &outputs));
+// 获取 outputs ...
+```
+
+编译时，联编 `libpaddle_fluid.a/.so` 和 `libpaddle_inference_api.a/.so` 便可。 
+
+## 详细代码参考
+
+- [inference demos](./demo)
+- [复杂单线程/多线程例子](https://github.com/PaddlePaddle/Paddle/blob/develop/paddle/contrib/inference/test_paddle_inference_api_impl.cc)
diff --git a/paddle/contrib/inference/paddle_inference_api.h b/paddle/contrib/inference/paddle_inference_api.h
index bd4530fcf9518cb3bf06179d8f60a1dde38ff7dd..38e3cc21413b9ab715b84f278f00b9df23cb7682 100644
--- a/paddle/contrib/inference/paddle_inference_api.h
+++ b/paddle/contrib/inference/paddle_inference_api.h
@@ -109,8 +109,7 @@ class PaddlePredictor {
 
   // The common configs for all the predictors.
   struct Config {
-    std::string model_dir;      // path to the model directory.
-    bool enable_engine{false};  // Enable to execute (part of) the model on
+    std::string model_dir;  // path to the model directory.
   };
 };