diff --git a/mindspore/lite/nnacl/fp32/tile.h b/mindspore/lite/nnacl/fp32/tile.h
index 967b7ff524..3c2de3fe0d 100644
--- a/mindspore/lite/nnacl/fp32/tile.h
+++ b/mindspore/lite/nnacl/fp32/tile.h
@@ -24,6 +24,7 @@ typedef struct TileParameter {
   int in_dim_;
   int in_shape_[5];
   int out_shape_[5];
+  int dims_[5];
   int multiples_[5];
   int in_strides_[5];
   int out_strides_[5];
diff --git a/mindspore/lite/src/ops/populate/tile_populate.cc b/mindspore/lite/src/ops/populate/tile_populate.cc
index d745331483..8d8bbdd2b5 100644
--- a/mindspore/lite/src/ops/populate/tile_populate.cc
+++ b/mindspore/lite/src/ops/populate/tile_populate.cc
@@ -31,10 +31,13 @@ OpParameter *PopulateTileParameter(const mindspore::lite::PrimitiveC *primitive)
   memset(tile_param, 0, sizeof(TileParameter));
   tile_param->op_parameter_.type_ = primitive->Type();
   auto param = reinterpret_cast<mindspore::lite::Tile *>(const_cast<mindspore::lite::PrimitiveC *>(primitive));
+  auto dims = param->GetDims();
   auto multiples = param->GetMultiples();
-  tile_param->in_dim_ = multiples.size();
-  for (int i = 0; i < tile_param->in_dim_; ++i) {
-    tile_param->multiples_[i] = multiples[i];
+  for (size_t i = 0; i < kDimension_4d; ++i) {
+    tile_param->multiples_[i] = 1;
+  }
+  for (size_t i = 0; i < dims.size(); ++i) {
+    tile_param->multiples_[dims[i]] = multiples[i];
   }
   return reinterpret_cast<OpParameter *>(tile_param);
 }
diff --git a/mindspore/lite/src/ops/tile.cc b/mindspore/lite/src/ops/tile.cc
index 49e907c9ab..8b18d709fe 100644
--- a/mindspore/lite/src/ops/tile.cc
+++ b/mindspore/lite/src/ops/tile.cc
@@ -140,18 +140,17 @@ int Tile::InferShape(std::vector<Tensor *> inputs_, std::vector<Tensor *> output
 
   std::vector<int> out_shape;
   std::vector<int> multiples = GetMultiples();
+  std::vector<int> dims = GetDims();
   const size_t in_dims = input->shape().size();
-  const size_t delta_dims = in_dims - multiples.size();
 
-  size_t i = 0;
-  for (; i < delta_dims; ++i) {
-    int tmp = input->shape()[i];
-    out_shape.push_back(tmp);
+  MS_ASSERT(multiples.size() == dims.size());
+  for (size_t i = 0; i < in_dims; ++i) {
+    out_shape.push_back(input->shape()[i]);
   }
-  for (; i < in_dims; ++i) {
-    int tmp = input->shape()[i] * (multiples[i - delta_dims]);
-    out_shape.push_back(tmp);
+  for (size_t i = 0; i < dims.size(); ++i) {
+    out_shape[dims[i]] = input->shape()[dims[i]] * (multiples[i]);
   }
+
   output->set_shape(out_shape);
   return RET_OK;
 }
diff --git a/mindspore/lite/src/runtime/kernel/arm/fp32/tile_fp32.cc b/mindspore/lite/src/runtime/kernel/arm/fp32/tile_fp32.cc
index 76aaff4422..3b212ea887 100644
--- a/mindspore/lite/src/runtime/kernel/arm/fp32/tile_fp32.cc
+++ b/mindspore/lite/src/runtime/kernel/arm/fp32/tile_fp32.cc
@@ -41,6 +41,7 @@ void TileCPUKernel::ComputeStrides(const int *shape, int *strides, int ndim) {
 
 int TileCPUKernel::ReSize() {
   auto tile_parameter_ = reinterpret_cast<TileParameter *>(op_parameter_);
+  tile_parameter_->in_dim_ = in_tensors_[0]->shape().size();
   for (int i = 0; i < tile_parameter_->in_dim_; ++i) {
     tile_parameter_->in_shape_[i] = in_tensors_[0]->shape()[i];
     tile_parameter_->out_shape_[i] = out_tensors_[0]->shape()[i];