[onert] Support to load BROADCAST_TO operation

runtime/onert/core/src/loader/BaseLoader.h

@@ -1721,6 +1721,9 @@ void BaseLoader<LoaderDomain>::loadOperation(const Operator *op, ir::Graph &subg
     case BuiltinOperator::BuiltinOperator_BATCH_MATMUL:
       loadBatchMatMul(op, subg);
       return;
+    case BuiltinOperator::BuiltinOperator_BROADCAST_TO:
+      loadOperationTo<ir::operation::BroadcastTo>(op, subg);
+      return;
     case BuiltinOperator::BuiltinOperator_LOG_SOFTMAX:
       loadLogSoftmax(op, subg);
       return;

runtime/tests/nnfw_api/lib/CircleGen.cc

@@ -148,6 +148,13 @@ uint32_t CircleGen::addOperatorBatchMatMul(const OperatorParams &params, bool ad
                                 circle::BuiltinOptions_BatchMatMulOptions, options);
 }
 
+uint32_t CircleGen::addOperatorBroadcastTo(const OperatorParams &params)
+{
+  auto options = circle::CreateBroadcastToOptions(_fbb).Union();
+  return addOperatorWithOptions(params, circle::BuiltinOperator_BROADCAST_TO,
+                                circle::BuiltinOptions_BroadcastToOptions, options);
+}
+
 uint32_t CircleGen::addOperatorCast(const OperatorParams &params, circle::TensorType input_type,
                                     circle::TensorType output_type)
 {

runtime/tests/nnfw_api/lib/CircleGen.h

@@ -150,6 +150,7 @@ class CircleGen
   uint32_t addOperatorBatchMatMul(const OperatorParams &params, bool adj_x, bool adj_y,
                                   bool asymmetric_quantize_inputs = false);
   uint32_t addOperatorBatchToSpaceND(const OperatorParams &params);
+  uint32_t addOperatorBroadcastTo(const OperatorParams &params);
   uint32_t addOperatorCast(const OperatorParams &params, circle::TensorType input_type,
                            circle::TensorType output_type);
   uint32_t addOperatorConcatenation(const OperatorParams &params, int axis,

runtime/tests/nnfw_api/src/GenModelTests/one_op_tests/BroadcastTo.test.cc

@@ -0,0 +1,68 @@
+/*
+ * Copyright (c) 2025 Samsung Electronics Co., Ltd. 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.
+ */
+
+#include "GenModelTest.h"
+
+TEST_F(GenModelTest, OneOp_BroadcastTo_1D_to_2D)
+{
+  CircleGen cgen;
+  const uint32_t shape_buf = cgen.addBuffer(std::vector<int32_t>{3, 3});
+  int shape = cgen.addTensor({{2}, circle::TensorType::TensorType_INT32, shape_buf});
+  int in = cgen.addTensor({{3}, circle::TensorType::TensorType_FLOAT32});
+  int out = cgen.addTensor({{3, 3}, circle::TensorType::TensorType_FLOAT32});
+  cgen.addOperatorBroadcastTo({{in, shape}, {out}});
+  cgen.setInputsAndOutputs({in}, {out});
+
+  _context = std::make_unique<GenModelTestContext>(cgen.finish());
+  _context->addTestCase(uniformTCD<float>({{1, 2, 3}}, {{1, 2, 3, 1, 2, 3, 1, 2, 3}}));
+  _context->setBackends({"cpu"});
+
+  SUCCEED();
+}
+
+TEST_F(GenModelTest, OneOp_BroadcastTo_2D_to_3D)
+{
+  CircleGen cgen;
+  const uint32_t shape_buf = cgen.addBuffer(std::vector<int32_t>{3, 2, 2});
+  int shape = cgen.addTensor({{3}, circle::TensorType::TensorType_INT32, shape_buf});
+  int in = cgen.addTensor({{2, 2}, circle::TensorType::TensorType_FLOAT32});
+  int out = cgen.addTensor({{3, 2, 2}, circle::TensorType::TensorType_FLOAT32});
+  cgen.addOperatorBroadcastTo({{in, shape}, {out}});
+  cgen.setInputsAndOutputs({in}, {out});
+
+  _context = std::make_unique<GenModelTestContext>(cgen.finish());
+  _context->addTestCase(uniformTCD<float>({{1, 2, 3, 4}}, {{1, 2, 3, 4, 1, 2, 3, 4, 1, 2, 3, 4}}));
+  _context->setBackends({"cpu"});
+
+  SUCCEED();
+}
+
+TEST_F(GenModelTest, OneOp_BroadcastTo_3D_to_3D)
+{
+  CircleGen cgen;
+  const uint32_t shape_buf = cgen.addBuffer(std::vector<int32_t>{2, 3, 2});
+  int shape = cgen.addTensor({{3}, circle::TensorType::TensorType_INT32, shape_buf});
+  int in = cgen.addTensor({{2, 1, 2}, circle::TensorType::TensorType_FLOAT32});
+  int out = cgen.addTensor({{2, 3, 2}, circle::TensorType::TensorType_FLOAT32});
+  cgen.addOperatorBroadcastTo({{in, shape}, {out}});
+  cgen.setInputsAndOutputs({in}, {out});
+
+  _context = std::make_unique<GenModelTestContext>(cgen.finish());
+  _context->addTestCase(uniformTCD<float>({{1, 2, 3, 4}}, {{1, 2, 1, 2, 1, 2, 3, 4, 3, 4, 3, 4}}));
+  _context->setBackends({"cpu"});
+
+  SUCCEED();
+}