diff --git a/tensorflow/compiler/mlir/tensorflow/BUILD b/tensorflow/compiler/mlir/tensorflow/BUILD
index dbe68a10099..4b01c8d0655 100644
--- a/tensorflow/compiler/mlir/tensorflow/BUILD
+++ b/tensorflow/compiler/mlir/tensorflow/BUILD
@@ -255,6 +255,7 @@ cc_library(
         "transforms/shape_inference_pass.cc",
         "transforms/sink_constant.cc",
         "transforms/test_side_effect_analysis.cc",
+        "transforms/tf_device_assignment.cc",
         "transforms/tpu_cluster_formation.cc",
         "transforms/tpu_dynamic_layout_pass.cc",
         "transforms/tpu_dynamic_padding_mapper.cc",
diff --git a/tensorflow/compiler/mlir/tensorflow/tests/device_assignment.mlir b/tensorflow/compiler/mlir/tensorflow/tests/device_assignment.mlir
new file mode 100644
index 00000000000..1f1e6c63f30
--- /dev/null
+++ b/tensorflow/compiler/mlir/tensorflow/tests/device_assignment.mlir
@@ -0,0 +1,13 @@
+// RUN: tf-opt -tf-simple-device-assignment='default-device=gpu' %s | FileCheck %s
+
+// CHECK-LABEL: func @device_test
+func @device_test(%arg0: tensor<3x1xf32>) -> (tensor<3x3xf32>) {
+
+  // CHECK: device = "gpu"
+  %0 = "tf.Const"() {value = dense<[[1.0, 2.0, 3.0]]> : tensor<1x3xf32>} : () -> tensor<1x3xf32>
+  // CHECK: device = "gpu"
+  %1 = "tf.MatMul"(%arg0, %0) {T = f32, _output_shapes = ["tfshape$dim { size: 3 } dim { size: 3 }"], device = "", transpose_a = false, transpose_b = false} : (tensor<3x1xf32>, tensor<1x3xf32>) -> tensor<3x3xf32>
+  // CHECK: device = "cpu"
+  %2 = "tf.Relu"(%1) {T = f32, _output_shapes = ["tfshape$dim { size: 3 } dim { size: 3 }"], device = "cpu"} : (tensor<3x3xf32>) -> tensor<3x3xf32>
+  return %2 : tensor<3x3xf32>
+}
diff --git a/tensorflow/compiler/mlir/tensorflow/transforms/passes.h b/tensorflow/compiler/mlir/tensorflow/transforms/passes.h
index 740dca71710..ff8f571f7d6 100644
--- a/tensorflow/compiler/mlir/tensorflow/transforms/passes.h
+++ b/tensorflow/compiler/mlir/tensorflow/transforms/passes.h
@@ -79,6 +79,10 @@ LogicalResult MarkFunctionVisibilityUsingEntryFunctionSpecification(
 std::unique_ptr<OpPassBase<ModuleOp>>
 CreateMarkFunctionVisibilityUsingEntryFunctionSpecificationPass();
 
+// Create a simple device assignment pass on TF dialect for CoreRT use case.
+std::unique_ptr<OpPassBase<FuncOp>> CreateSimpleTFDeviceAssignmentPass(
+    llvm::StringRef default_device);
+
 }  // namespace TF
 
 namespace TFControlFlow {
diff --git a/tensorflow/compiler/mlir/tensorflow/transforms/tf_device_assignment.cc b/tensorflow/compiler/mlir/tensorflow/transforms/tf_device_assignment.cc
new file mode 100644
index 00000000000..a4a8c1ab95f
--- /dev/null
+++ b/tensorflow/compiler/mlir/tensorflow/transforms/tf_device_assignment.cc
@@ -0,0 +1,68 @@
+/* Copyright 2019 The TensorFlow Authors. All Rights Reserved.
+
+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.
+==============================================================================*/
+
+// This file implements device assignment in TF dialect.
+#include "mlir/IR/Builders.h"
+#include "mlir/Pass/Pass.h"
+#include "tensorflow/compiler/mlir/tensorflow/ir/tf_ops.h"
+#include "tensorflow/compiler/mlir/tensorflow/transforms/passes.h"
+
+namespace mlir {
+namespace TF {
+namespace {
+
+class SimpleTFDeviceAssignmentPass
+    : public FunctionPass<SimpleTFDeviceAssignmentPass> {
+ public:
+  SimpleTFDeviceAssignmentPass() = default;
+  SimpleTFDeviceAssignmentPass(const SimpleTFDeviceAssignmentPass&) {}
+  explicit SimpleTFDeviceAssignmentPass(llvm::StringRef default_device) {
+    default_device_ = std::string(default_device);
+  }
+
+  void runOnFunction() override {
+    Builder builder(&getContext());
+    getFunction().walk([this, &builder](Operation* op) {
+      if (auto device_attr = op->getAttrOfType<StringAttr>("device")) {
+        // We assign default device to ops with device attribute that is empty.
+        if (device_attr.getValue() == "") {
+          op->setAttr("device", builder.getStringAttr(default_device_));
+        }
+      } else if (llvm::isa<ConstOp>(op)) {
+        // tf.Const may sometimes contain no device attribute. In this case, we
+        // assign it the default device.
+        op->setAttr("device", builder.getStringAttr(default_device_));
+      }
+    });
+  }
+
+ private:
+  Option<std::string> default_device_{
+      *this, "default-device", llvm::cl::desc("The default device to assign."),
+      llvm::cl::init("cpu")};
+};
+
+}  // namespace
+
+std::unique_ptr<OpPassBase<FuncOp>> CreateSimpleTFDeviceAssignmentPass(
+    llvm::StringRef default_device) {
+  return std::make_unique<SimpleTFDeviceAssignmentPass>(default_device);
+}
+
+static PassRegistration<SimpleTFDeviceAssignmentPass> pass(
+    "tf-simple-device-assignment", "Simple device assignment in TF dialect.");
+
+}  // namespace TF
+}  // namespace mlir