diff --git a/cinn/frontend/net_builder_test.cc b/cinn/frontend/net_builder_test.cc
index 554b308d79..717ecb6a61 100755
--- a/cinn/frontend/net_builder_test.cc
+++ b/cinn/frontend/net_builder_test.cc
@@ -269,14 +269,12 @@ TEST(net_build, program_execute_gather) {
   scope->Var<hlir::framework::Tensor>(std::string(output->id));
 
   auto input1_tensor = scope->GetTensor(std::string(input1.id()));
-  //  SetRandData<float>(input1_tensor, target);
+  SetRandData<float>(input1_tensor, target);
   float* input1_data = input1_tensor->mutable_data<float>(target);
 
   auto input2_tensor = scope->GetTensor(std::string(input2.id()));
-  //  SetRandData<int>(input2_tensor, target);
+  SetRandInt(input2_tensor, target);
   int* input2_data = input2_tensor->mutable_data<int>(target);
-
-  memset(input1_data, 0, sizeof(float) * B * H_IN1);
   memset(input2_data, 0, sizeof(int) * B * H_IN2);
   runtime_program->Execute();
 
@@ -329,7 +327,7 @@ TEST(net_build, program_execute_gather_nd) {
   float* input1_data = input1_tensor->mutable_data<float>(target);
 
   auto input2_tensor = scope->GetTensor(std::string(input2.id()));
-  SetRandData<int>(input2_tensor, target);
+  SetRandInt(input2_tensor, target);
   int* input2_data = input2_tensor->mutable_data<int>(target);
 
   runtime_program->Execute();
@@ -380,11 +378,11 @@ TEST(net_build, program_execute_scatter) {
   scope->Var<hlir::framework::Tensor>(std::string(output->id));
 
   auto input1_tensor = scope->GetTensor(std::string(input1.id()));
-  //  SetRandData<float>(input1_tensor, target);
+  SetRandData<float>(input1_tensor, target);
   float* input1_data = input1_tensor->mutable_data<float>(target);
 
   auto input2_tensor = scope->GetTensor(std::string(input2.id()));
-  //  SetRandData<int>(input2_tensor, target);
+  //  SetRandInt(input2_tensor, target);
   int* input2_data = input2_tensor->mutable_data<int>(target);
 
   runtime_program->Execute();
@@ -452,7 +450,7 @@ TEST(net_build, program_execute_scatter_nd) {
   SetRandData<float>(input1_tensor, target);
 
   auto input2_tensor = scope->GetTensor(std::string(input2.id()));
-  SetRandData<int>(input2_tensor, target);
+  SetRandInt(input2_tensor, target);
 
   runtime_program->Execute();
 
@@ -517,7 +515,7 @@ TEST(net_build, program_execute_cast) {
   scope->Var<hlir::framework::Tensor>(std::string(output->id));
 
   auto input_tensor = scope->GetTensor(std::string(input.id()));
-  SetRandData<int>(input_tensor, target);
+  SetRandInt(input_tensor, target);
   int* input_data = input_tensor->mutable_data<int>(target);
 
   runtime_program->Execute();
diff --git a/cinn/hlir/op/contrib/gather.cc b/cinn/hlir/op/contrib/gather.cc
index 9a566f4ccd..bef7a7dd18 100644
--- a/cinn/hlir/op/contrib/gather.cc
+++ b/cinn/hlir/op/contrib/gather.cc
@@ -55,7 +55,7 @@ ir::Tensor Gather(const ir::Tensor &A, const ir::Tensor &B, const int &axis, con
         for (int i = 0; i < axis; ++i) {
           A_indices.push_back(indices[i]);
         }
-        A_indices.push_back(ir::Cast::Make(Int(32), B(indices)));
+        A_indices.push_back(B(indices));
         for (size_t i = axis + 1; i < A->shape.size(); ++i) {
           A_indices.push_back(indices[i]);
         }
diff --git a/cinn/utils/data_util.cc b/cinn/utils/data_util.cc
index b9cbc339aa..96d01b2dc0 100644
--- a/cinn/utils/data_util.cc
+++ b/cinn/utils/data_util.cc
@@ -14,10 +14,11 @@
 
 #include "cinn/utils/data_util.h"
 
+#include "iostream"
+
 namespace cinn {
 
-template <typename T>
-void SetRandInt<int>(hlir::framework::Tensor tensor, const common::Target& target, int seed) {
+void SetRandInt(hlir::framework::Tensor tensor, const common::Target& target, int seed) {
   if (seed == -1) {
     std::random_device rd;
     seed = rd();
@@ -25,20 +26,21 @@ void SetRandInt<int>(hlir::framework::Tensor tensor, const common::Target& targe
   std::default_random_engine engine(seed);
   std::uniform_int_distribution<int> dist(1, 10);
   size_t num_ele = tensor->shape().numel();
-  std::vector<T> random_data(num_ele);
+  std::vector<int> random_data(num_ele);
   for (size_t i = 0; i < num_ele; i++) {
-    random_data[i] = static_cast<T> dist(engine);  // All random data
+    random_data[i] = static_cast<int>(dist(engine));  // All random data
   }
 
-  auto* data = tensor->mutable_data<T>(target);
+  auto* data = tensor->mutable_data<int>(target);
 #ifdef CINN_WITH_CUDA
   if (target == common::DefaultNVGPUTarget()) {
-    cudaMemcpy(data, random_data.data(), num_ele * sizeof(T), cudaMemcpyHostToDevice);
+    cudaMemcpy(data, random_data.data(), num_ele * sizeof(int), cudaMemcpyHostToDevice);
     return;
   }
 #endif
   CHECK(target == common::DefaultHostTarget());
   std::copy(random_data.begin(), random_data.end(), data);
+  std::cout << "success" << std::endl;
 }
 
 template <>
diff --git a/cinn/utils/data_util.h b/cinn/utils/data_util.h
index 57ad14a916..d1d765c549 100644
--- a/cinn/utils/data_util.h
+++ b/cinn/utils/data_util.h
@@ -23,7 +23,6 @@
 #endif
 
 namespace cinn {
-template <typename T>
 void SetRandInt(hlir::framework::Tensor tensor, const common::Target& target, int seed = -1);
 
 template <typename T>