diff --git a/paddle/fluid/inference/api/api_anakin_engine.cc b/paddle/fluid/inference/api/api_anakin_engine.cc
index 86e6fdaabf2dd4..4c51c239f6d444 100644
--- a/paddle/fluid/inference/api/api_anakin_engine.cc
+++ b/paddle/fluid/inference/api/api_anakin_engine.cc
@@ -208,6 +208,7 @@ bool PaddleInferenceAnakinPredictor<T, P, R>::RunImpl(
       LOG(FATAL) << " input  " << input.name
                  << "'s shape size should be equal to that of net";
     }
+#ifndef ANAKIN_MLU_PLACE
     int sum = 1;
     for_each(input.shape.begin(), input.shape.end(), [&](int n) { sum *= n; });
     if (sum > net_shape.count()) {
@@ -221,6 +222,7 @@ bool PaddleInferenceAnakinPredictor<T, P, R>::RunImpl(
                "memory.";
       }
     }
+#endif
     std::vector<int> tmp_shape;
     for (auto s : input.shape) {
       tmp_shape.push_back(s);
@@ -229,8 +231,9 @@ bool PaddleInferenceAnakinPredictor<T, P, R>::RunImpl(
     anakin::saber::Tensor<typename anakin::DefaultHostType<T>::Host_type>
         h_tensor(data, typename anakin::DefaultHostType<T>::Host_type(), 0,
                  tmp_shape);
+#ifndef ANAKIN_MLU_PLACE
     d_tensor_p->reshape(tmp_shape);
-
+#endif
     if (input.lod.size() > 0) {
       if (input.lod.size() > 1) {
         LOG(FATAL) << " input lod first dim should <=1, but you set "
@@ -256,14 +259,18 @@ bool PaddleInferenceAnakinPredictor<T, P, R>::RunImpl(
       LOG(FATAL) << output.name << " is not in the outputs of the graph.";
     }
     auto *d_tensor_p = this->executor_p_->get_out(output.name);
-    output.shape = d_tensor_p->valid_shape();
-    if (output.data.length() < d_tensor_p->valid_size() * sizeof(float)) {
-      output.data.Resize(d_tensor_p->valid_size() * sizeof(float));
+    auto tmp_shape = d_tensor_p->valid_shape();
+#ifdef ANAKIN_MLU_PLACE
+    tmp_shape.set_num(batch_size);
+#endif
+    output.shape = tmp_shape;
+    if (output.data.length() < tmp_shape.count() * sizeof(float)) {
+      output.data.Resize(tmp_shape.count() * sizeof(float));
     }
     auto *data = static_cast<float *>(output.data.data());
     anakin::saber::Tensor<typename anakin::DefaultHostType<T>::Host_type>
         h_tensor(data, typename anakin::DefaultHostType<T>::Host_type(), 0,
-                 d_tensor_p->valid_shape());
+                 tmp_shape);
     h_tensor.copy_from(*d_tensor_p);
   }
   return true;
@@ -316,6 +323,7 @@ void PaddleInferenceAnakinMLUPredictor<P, R>::SetContext() {
       this->config_.device_id, this->config_.data_stream_id,
       this->config_.compute_stream_id);
   this->ctx_p_->set_model_parallel(this->config_.model_parallel);
+  this->ctx_p_->set_fusion(this->config_.op_fuse);
   this->ctx_p_->enable_batch_changable();
   this->ctx_p_->enable_channel_duplicate();
 }