diff --git a/paddle/fluid/inference/api/paddle_api.h b/paddle/fluid/inference/api/paddle_api.h
index d9edcf7cc5eefbd64883b679b39838f3f0e6a993..46b510fd1ec94c59032b8f41a2ac4d6aa87dc150 100644
--- a/paddle/fluid/inference/api/paddle_api.h
+++ b/paddle/fluid/inference/api/paddle_api.h
@@ -170,18 +170,40 @@ class PaddlePredictor {
                    std::vector<PaddleTensor>* output_data,
                    int batch_size = -1) = 0;
 
-  /** Zero copy input and output optimization.
-   * Get the input or output tensors, and operate on their memory directly,
-   * without copy.
+  /** \brief Get a mutable tensor directly.
+   *
+   * NOTE Only works in AnalysisPredictor.
+   *
+   * One can also use this to modify any temporary variable related tensors in
+   * the predictor.
+   *
    */
   virtual std::unique_ptr<ZeroCopyTensor> GetInputTensor(
       const std::string& name) {
     return nullptr;
   }
+  /**
+   * \brief Get an immutable tensor without copy.
+   *
+   * NOTE Only works in AnalysisPredictor.
+   * One can use this API to get any temporary tensors in the predictor and
+   * read it.
+   */
   virtual std::unique_ptr<ZeroCopyTensor> GetOutputTensor(
       const std::string& name) {
     return nullptr;
   }
+  /**
+   * \brief Run the predictor with zero-copied inputs and outputs.
+   *
+   * NOTE Only works in AnalysisPredictor.
+   *
+   * This will save the IO copy for transfering inputs and outputs to predictor
+   * workspace and get some performance improvement.
+   * To use it, one should call the `AnalysisConfig.SwitchUseFeedFetchOp(true)`
+   * and then use the `GetInputTensor` and `GetOutputTensor` to directly write
+   * or read the input/output tensors.
+   */
   virtual bool ZeroCopyRun() { return false; }
 
   /** Clone a predictor that share the model weights, the Cloned predictor