diff --git a/mindspore/ops/_op_impl/tbe/__init__.py b/mindspore/ops/_op_impl/tbe/__init__.py
index 9e7c8745ac..c18a0a28c3 100644
--- a/mindspore/ops/_op_impl/tbe/__init__.py
+++ b/mindspore/ops/_op_impl/tbe/__init__.py
@@ -125,6 +125,8 @@ from .layer_norm import _layer_norm_tbe
 from .layer_norm_grad import _layer_norm_grad_tbe
 from .layer_norm_x_backprop import _layer_norm_x_backprop_tbe
 from .l2_loss import _l2_loss_tbe
+from .l2_normalize import _l2_normalize_tbe
+from .l2_normalize_grad import _l2_normalize_grad_tbe
 from .square_sum_v1 import _square_sum_v1_tbe
 from .square_sum_v2 import _square_sum_v2_tbe
 from .confusion_transpose_d import _confusion_transpose_d_tbe
diff --git a/mindspore/ops/_op_impl/tbe/l2_normalize.py b/mindspore/ops/_op_impl/tbe/l2_normalize.py
new file mode 100644
index 0000000000..9790592d9a
--- /dev/null
+++ b/mindspore/ops/_op_impl/tbe/l2_normalize.py
@@ -0,0 +1,38 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+
+"""L2Normalize op"""
+from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
+
+l2_normalize_op_info = TBERegOp("L2Normalize") \
+    .fusion_type("OPAQUE") \
+    .async_flag(False) \
+    .binfile_name("l2_normalize.so") \
+    .compute_cost(10) \
+    .kernel_name("l2_normalize") \
+    .partial_flag(True) \
+    .attr("axis", "required", "listInt", "all") \
+    .attr("epsilon", "required", "float", "all") \
+    .input(0, "x", False, "required", "all") \
+    .output(0, "y", True, "required", "all") \
+    .dtype_format(DataType.F16_Default, DataType.F16_Default) \
+    .dtype_format(DataType.F32_Default, DataType.F32_Default) \
+    .get_op_info()
+
+
+@op_info_register(l2_normalize_op_info)
+def _l2_normalize_tbe():
+    """L2Normalize TBE register"""
+    return
diff --git a/mindspore/ops/_op_impl/tbe/l2_normalize_grad.py b/mindspore/ops/_op_impl/tbe/l2_normalize_grad.py
new file mode 100644
index 0000000000..e164120c75
--- /dev/null
+++ b/mindspore/ops/_op_impl/tbe/l2_normalize_grad.py
@@ -0,0 +1,40 @@
+# Copyright 2020 Huawei Technologies Co., Ltd
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ============================================================================
+
+"""L2NormalizeGrad op"""
+from mindspore.ops.op_info_register import op_info_register, TBERegOp, DataType
+
+l2_normalize_grad_op_info = TBERegOp("L2NormalizeGrad") \
+    .fusion_type("OPAQUE") \
+    .async_flag(False) \
+    .binfile_name("l2_normalize_grad.so") \
+    .compute_cost(10) \
+    .kernel_name("l2_normalize_grad") \
+    .partial_flag(True) \
+    .attr("axis", "required", "listInt", "all") \
+    .attr("epsilon", "required", "float", "all") \
+    .input(0, "x", False, "required", "all") \
+    .input(1, "y", False, "required", "all") \
+    .input(2, "dy", False, "requried", "all") \
+    .output(0, "dx", True, "required", "all") \
+    .dtype_format(DataType.F16_Default, DataType.F16_Default, DataType.F16_Default, DataType.F16_Default) \
+    .dtype_format(DataType.F32_Default, DataType.F32_Default, DataType.F32_Default, DataType.F32_Default) \
+    .get_op_info()
+
+
+@op_info_register(l2_normalize_grad_op_info)
+def _l2_normalize_grad_tbe():
+    """L2NormalizeGrad TBE register"""
+    return
diff --git a/tests/ut/python/ops/test_nn_ops.py b/tests/ut/python/ops/test_nn_ops.py
index 5038ee28a0..0e9b8d2afa 100644
--- a/tests/ut/python/ops/test_nn_ops.py
+++ b/tests/ut/python/ops/test_nn_ops.py
@@ -198,6 +198,18 @@ class ScalarSummaryNet(nn.Cell):
         return out
 
 
+class L2NormalizeNet(nn.Cell):
+    """ L2NormalizeNet definition """
+
+    def __init__(self):
+        super(L2NormalizeNet, self).__init__()
+        self.l2_normalize = P.L2Normalize()
+
+    def construct(self, x):
+        out = self.l2_normalize(x)
+        return out
+
+
 class HistogramSummaryNet(nn.Cell):
     """HistogramSummaryNet definition"""
 
@@ -450,6 +462,10 @@ test_cases = [
         'block': ScalarSummaryNet(),
         'desc_inputs': [2.2],
     }),
+    ('L2Normalize', {
+        'block': L2NormalizeNet(),
+        'desc_inputs': [Tensor(np.array([[1.0, 2, 3], [4.0, 5, 6], [7.0, 8, 9]]), mindspore.float32)],
+    }),
     ('HistogramSummary', {
         'block': HistogramSummaryNet(),
         'desc_inputs': [[1,2,3]],
diff --git a/tests/ut/python/ops/test_ops.py b/tests/ut/python/ops/test_ops.py
index 1c1d9d5335..058df1a0f1 100755
--- a/tests/ut/python/ops/test_ops.py
+++ b/tests/ut/python/ops/test_ops.py
@@ -681,6 +681,15 @@ test_case_nn_ops = [
         'desc_inputs': [[16, 1234], [16, 1234]],
         'desc_bprop': [[64, 2]],
         'skip': ['backward']}),
+    ('L2Normalize', {
+        'block': P.L2Normalize(),
+        'desc_inputs': [[2, 2]],
+        'desc_bprop': [[2, 2]]}),
+    ('L2NormalizeGrad', {
+        'block': G.L2NormalizeGrad(),
+        'desc_inputs': [[2, 2], [2, 2], [2, 2]],
+        'desc_bprop': [[2, 2]],
+        'skip': ['backward']}),
     ('LayerNorm', {
         'block': P.LayerNorm(),
         'desc_inputs': [[2, 16], [16], [16]],