diff --git a/Dockerfile b/Dockerfile
index fe0721e9b99b5e028df2f6228ff04cb56a567a3f..f5cc824c4172bf8cadf5168374994c0e3a658329 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -75,7 +75,8 @@ RUN curl -s -q https://glide.sh/get | sh
 #    and its size is only one-third of the official one.
 # 2. Manually add ~IPluginFactory() in IPluginFactory class of NvInfer.h, otherwise, it couldn't work in paddle.
 #    See https://github.com/PaddlePaddle/Paddle/issues/10129 for details.
-RUN wget -qO- http://paddlepaddledeps.cdn.bcebos.com/TensorRT-4.0.0.3.Ubuntu-16.04.4.x86_64-gnu.cuda-8.0.cudnn7.0.tar.gz | \
+
+RUN wget -qO- https://paddlepaddledeps.cdn.bcebos.com/TensorRT-4.0.1.6-ubuntu14.04.x86_64-gnu.cuda-8.0.cudnn7.0.tar.gz | \
     tar -xz -C /usr/local && \
     cp -rf /usr/local/TensorRT/include /usr && \
     cp -rf /usr/local/TensorRT/lib /usr
diff --git a/paddle/fluid/inference/api/analysis_predictor.cc b/paddle/fluid/inference/api/analysis_predictor.cc
index 467d4411376381df950bb582f9c73410284a5e2d..8020827d30f9324361064098b5cdb953b30ea9ac 100644
--- a/paddle/fluid/inference/api/analysis_predictor.cc
+++ b/paddle/fluid/inference/api/analysis_predictor.cc
@@ -438,12 +438,14 @@ void AnalysisPredictor::PrepareFeedFetch() {
       }
       feeds_[idx] = op;
       feed_names_[op->Output("Out")[0]] = idx;
+      idx2feeds_[idx] = op->Output("Out")[0];
     } else if (op->Type() == "fetch") {
       int idx = boost::get<int>(op->GetAttr("col"));
       if (fetches_.size() <= static_cast<size_t>(idx)) {
         fetches_.resize(idx + 1);
       }
       fetches_[idx] = op;
+      idx2fetches_[idx] = op->Input("X")[0];
     }
   }
 }
@@ -456,6 +458,22 @@ void AnalysisPredictor::CreateFeedFetchVar(framework::Scope *scope) {
   var->GetMutable<framework::FeedFetchList>();
 }
 
+std::vector<std::string> AnalysisPredictor::GetInputNames() {
+  std::vector<std::string> input_names;
+  for (auto &item : idx2feeds_) {
+    input_names.push_back(item.second);
+  }
+  return input_names;
+}
+
+std::vector<std::string> AnalysisPredictor::GetOutputNames() {
+  std::vector<std::string> output_names;
+  for (auto &item : idx2fetches_) {
+    output_names.push_back(item.second);
+  }
+  return output_names;
+}
+
 std::unique_ptr<ZeroCopyTensor> AnalysisPredictor::GetInputTensor(
     const std::string &name) {
   PADDLE_ENFORCE(executor_->scope()->FindVar(name), "no name called %s", name);
@@ -463,6 +481,13 @@ std::unique_ptr<ZeroCopyTensor> AnalysisPredictor::GetInputTensor(
       new ZeroCopyTensor(static_cast<void *>(executor_->scope())));
   res->input_or_output_ = true;
   res->SetName(name);
+  if (platform::is_cpu_place(place_)) {
+    res->SetPlace(PaddlePlace::kCPU);
+  } else {
+    auto gpu_place = boost::get<platform::CUDAPlace>(place_);
+    res->SetPlace(PaddlePlace::kGPU, gpu_place.GetDeviceId());
+  }
+
   return res;
 }
 
@@ -473,6 +498,12 @@ std::unique_ptr<ZeroCopyTensor> AnalysisPredictor::GetOutputTensor(
       new ZeroCopyTensor(static_cast<void *>(executor_->scope())));
   res->input_or_output_ = false;
   res->SetName(name);
+  if (platform::is_cpu_place(place_)) {
+    res->SetPlace(PaddlePlace::kCPU);
+  } else {
+    auto gpu_place = boost::get<platform::CUDAPlace>(place_);
+    res->SetPlace(PaddlePlace::kGPU, gpu_place.GetDeviceId());
+  }
   return res;
 }
 
diff --git a/paddle/fluid/inference/api/analysis_predictor.h b/paddle/fluid/inference/api/analysis_predictor.h
index cc06e3479c3cb396247bcd077d267dd507242a39..5c0535d63e00c32ef82aa6d804459542d7da3e50 100644
--- a/paddle/fluid/inference/api/analysis_predictor.h
+++ b/paddle/fluid/inference/api/analysis_predictor.h
@@ -55,6 +55,9 @@ class AnalysisPredictor : public PaddlePredictor {
            std::vector<PaddleTensor> *output_data,
            int batch_size = -1) override;
 
+  std::vector<std::string> GetInputNames();
+  std::vector<std::string> GetOutputNames();
+
   std::unique_ptr<ZeroCopyTensor> GetInputTensor(
       const std::string &name) override;
   std::unique_ptr<ZeroCopyTensor> GetOutputTensor(
@@ -133,7 +136,11 @@ class AnalysisPredictor : public PaddlePredictor {
   std::shared_ptr<framework::ProgramDesc> inference_program_;
   std::vector<framework::OpDesc *> feeds_;
   std::map<std::string, size_t> feed_names_;
+  // Sorted according to the idx.
+  std::map<size_t, std::string> idx2feeds_;
   std::vector<framework::OpDesc *> fetches_;
+  std::map<size_t, std::string> idx2fetches_;
+
   // Memory buffer for feed inputs. The temporary LoDTensor will cause serious
   // concurrency problems, wrong results and memory leak, so cache them.
   std::vector<framework::LoDTensor> feed_tensors_;
diff --git a/paddle/fluid/inference/api/details/zero_copy_tensor.cc b/paddle/fluid/inference/api/details/zero_copy_tensor.cc
index f60ff40c5da3e9e03c2cb3583263394cb82db805..cf02901d963858d2a44b7c588a5c6a49358b0d3f 100644
--- a/paddle/fluid/inference/api/details/zero_copy_tensor.cc
+++ b/paddle/fluid/inference/api/details/zero_copy_tensor.cc
@@ -15,6 +15,7 @@
 #include "paddle/fluid/framework/lod_tensor.h"
 #include "paddle/fluid/framework/scope.h"
 #include "paddle/fluid/inference/api/paddle_inference_api.h"
+#include "paddle/fluid/memory/memcpy.h"
 #include "paddle/fluid/platform/enforce.h"
 
 namespace paddle {
@@ -73,6 +74,61 @@ T *ZeroCopyTensor::data(PaddlePlace *place, int *size) const {
   return res;
 }
 
+template <typename T>
+void ZeroCopyTensor::copy_from_cpu(const T *data) {
+  EAGER_GET_TENSOR;
+  PADDLE_ENFORCE_GE(
+      tensor->numel(), 0,
+      "You should call ZeroCopyTensor::Reshape(const std::vector<int> &shape)"
+      "function before copy data from cpu.");
+  size_t ele_size = tensor->numel() * sizeof(T);
+
+  if (place_ == PaddlePlace::kCPU) {
+    auto *t_data = tensor->mutable_data<T>(platform::CPUPlace());
+    std::memcpy(static_cast<void *>(t_data), data, ele_size);
+  } else {
+#ifdef PADDLE_WITH_CUDA
+    platform::DeviceContextPool &pool = platform::DeviceContextPool::Instance();
+    platform::CUDAPlace gpu_place(device_);
+    auto *t_data = tensor->mutable_data<T>(gpu_place);
+    auto *dev_ctx =
+        static_cast<const platform::CUDADeviceContext *>(pool.Get(gpu_place));
+
+    memory::Copy(gpu_place, static_cast<void *>(t_data), platform::CPUPlace(),
+                 data, ele_size, dev_ctx->stream());
+#else
+    PADDLE_THROW("Not compile with CUDA, should not reach here.");
+#endif
+  }
+}
+
+template <typename T>
+void ZeroCopyTensor::copy_to_cpu(T *data) {
+  EAGER_GET_TENSOR;
+  auto ele_num = tensor->numel();
+  auto *t_data = tensor->data<T>();
+  auto t_place = tensor->place();
+
+  if (platform::is_cpu_place(t_place)) {
+    std::memcpy(static_cast<void *>(data), t_data, ele_num * sizeof(T));
+  } else {
+#ifdef PADDLE_WITH_CUDA
+    platform::DeviceContextPool &pool = platform::DeviceContextPool::Instance();
+    auto gpu_place = boost::get<platform::CUDAPlace>(t_place);
+    auto *dev_ctx =
+        static_cast<const platform::CUDADeviceContext *>(pool.Get(gpu_place));
+    memory::Copy(platform::CPUPlace(), static_cast<void *>(data), gpu_place,
+                 t_data, ele_num * sizeof(T), dev_ctx->stream());
+#else
+    PADDLE_THROW("Not compile with CUDA, should not reach here.");
+#endif
+  }
+}
+template void ZeroCopyTensor::copy_from_cpu<float>(const float *data);
+template void ZeroCopyTensor::copy_from_cpu<int64_t>(const int64_t *data);
+template void ZeroCopyTensor::copy_to_cpu<float>(float *data);
+template void ZeroCopyTensor::copy_to_cpu<int64_t>(int64_t *data);
+
 template float *ZeroCopyTensor::data<float>(PaddlePlace *place,
                                             int *size) const;
 template int64_t *ZeroCopyTensor::data<int64_t>(PaddlePlace *place,
@@ -92,10 +148,10 @@ void *ZeroCopyTensor::FindTensor() const {
   return tensor;
 }
 
-std::vector<int64_t> ZeroCopyTensor::shape() const {
+std::vector<int> ZeroCopyTensor::shape() const {
   EAGER_GET_TENSOR;
   PADDLE_ENFORCE(tensor_, "not found tensor called %s in the scope", name_);
-  return framework::vectorize(tensor->dims());
+  return framework::vectorize2int(tensor->dims());
 }
 
 void ZeroCopyTensor::SetLoD(const std::vector<std::vector<size_t>> &x) {
diff --git a/paddle/fluid/inference/api/details/zero_copy_tensor_dummy.cc b/paddle/fluid/inference/api/details/zero_copy_tensor_dummy.cc
index 12071e09f8442f2c52a06b7c3fe4bed2c28b524a..cbbb3ea2d1395acdf4c460bea4b7868c31a20e53 100644
--- a/paddle/fluid/inference/api/details/zero_copy_tensor_dummy.cc
+++ b/paddle/fluid/inference/api/details/zero_copy_tensor_dummy.cc
@@ -37,7 +37,7 @@ template int64_t *ZeroCopyTensor::mutable_data(PaddlePlace place);
 
 void *ZeroCopyTensor::FindTensor() const { return nullptr; }
 
-std::vector<int64_t> ZeroCopyTensor::shape() const { return {}; }
+std::vector<int> ZeroCopyTensor::shape() const { return {}; }
 
 void ZeroCopyTensor::SetLoD(const std::vector<std::vector<size_t>> &x) {}
 
diff --git a/paddle/fluid/inference/api/paddle_api.h b/paddle/fluid/inference/api/paddle_api.h
index c9a45b4aa3b4037d3725622fc960848bc1ccfb2c..f807289f6aee06e3ff61bc0dd92f47c599421354 100644
--- a/paddle/fluid/inference/api/paddle_api.h
+++ b/paddle/fluid/inference/api/paddle_api.h
@@ -160,11 +160,21 @@ class ZeroCopyTensor {
   template <typename T>
   T* data(PaddlePlace* place, int* size) const;
 
-  std::vector<int64_t> shape() const;
+  template <typename T>
+  void copy_from_cpu(const T* data);
+
+  template <typename T>
+  void copy_to_cpu(T* data);
+
+  std::vector<int> shape() const;
 
   void SetLoD(const std::vector<std::vector<size_t>>& x);
   std::vector<std::vector<size_t>> lod() const;
   const std::string& name() const { return name_; }
+  void SetPlace(PaddlePlace place, int device = -1) {
+    place_ = place;
+    device_ = device;
+  }
 
  protected:
   explicit ZeroCopyTensor(void* scope) : scope_{scope} {}
@@ -179,6 +189,8 @@ class ZeroCopyTensor {
   // The corresponding tensor pointer inside Paddle workspace is cached for
   // performance.
   mutable void* tensor_{nullptr};
+  PaddlePlace place_;
+  int device_;
 };
 
 /** A simple Inference API for Paddle.
@@ -200,6 +212,14 @@ class PaddlePredictor {
                    std::vector<PaddleTensor>* output_data,
                    int batch_size = -1) = 0;
 
+  /** \brief Get input names of the model
+   */
+  virtual std::vector<std::string> GetInputNames() { return {}; }
+
+  /** \brief Get output names of the model
+   */
+  virtual std::vector<std::string> GetOutputNames() { return {}; }
+
   /** \brief Get a mutable tensor directly.
    *
    * NOTE Only works in AnalysisPredictor.