diff --git a/tests/CMakeLists.txt b/tests/CMakeLists.txt
index 9ccecbd075..c513c39a10 100644
--- a/tests/CMakeLists.txt
+++ b/tests/CMakeLists.txt
@@ -1,3 +1,5 @@
+cc_library(test_utils SRCS utils.cc DEPS cinncore)
+
 if (WITH_TESTING)
   include_directories(${CMAKE_SOURCE_DIR}/cinn/runtime)
   add_subdirectory(benchmark)
diff --git a/tests/utils.cc b/tests/utils.cc
new file mode 100644
index 0000000000..b3a9477c51
--- /dev/null
+++ b/tests/utils.cc
@@ -0,0 +1,126 @@
+// Copyright (c) 2022 CINN 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.
+
+#include "tests/utils.h"
+
+#include <glog/logging.h>
+
+#include "cinn/auto_schedule/analysis/analyze_ir.h"
+#include "cinn/backends/codegen_cuda_dev.h"
+#include "cinn/backends/compiler.h"
+#include "cinn/frontend/optimize.h"
+#include "cinn/hlir/framework/op_lowering.h"
+#ifdef CINN_WITH_CUDA
+#include <cuda_runtime.h>
+#endif
+
+namespace cinn {
+namespace tests {
+
+using ::cinn::hlir::framework::Instruction;
+using ::cinn::hlir::framework::Scope;
+using ::cinn::hlir::framework::Shape;
+using ::cinn::hlir::framework::Tensor;
+
+std::shared_ptr<hlir::framework::Graph> OptimizeByPass(frontend::Program& program, const common::Target& target) {
+  return frontend::Optimize(&program, {}, target);
+}
+
+std::vector<ir::LoweredFunc> LowerFusionGroup(std::shared_ptr<hlir::framework::Graph> graph,
+                                              std::shared_ptr<hlir::framework::Graph::Group> group,
+                                              const common::Target& target,
+                                              bool apply_manual_schedule) {
+  auto& dtype_dict = graph->GetMutableAttrs<absl::flat_hash_map<std::string, common::Type>>("inferdtype");
+  auto& shape_dict = graph->GetMutableAttrs<absl::flat_hash_map<std::string, hlir::framework::shape_t>>("infershape");
+  hlir::framework::OpLowerer op_lowerer(dtype_dict, shape_dict, target);
+
+  return apply_manual_schedule ? op_lowerer.Lower(group) : op_lowerer.LowerWithoutSchedule(group);
+}
+
+ir::IRSchedule MakeIRSchedule(const std::vector<ir::LoweredFunc>& lowered_funcs) {
+  std::vector<Expr> bodys;
+  for (auto&& func : lowered_funcs) {
+    bodys.emplace_back(func->body);
+  }
+  return ir::IRSchedule(ir::ModuleExpr({std::move(bodys)}));
+}
+
+std::vector<ir::LoweredFunc> OptimizeBySchedule(const ir::IRSchedule& schedule,
+                                                const std::vector<ir::LoweredFunc>& original_funcs,
+                                                const common::Target& target) {
+  auto&& updated_bodys = schedule.GetModule().GetExprs();
+  CHECK_EQ(updated_bodys.size(), original_funcs.size()) << "associated exprs size not equal";
+
+  std::vector<ir::LoweredFunc> results;
+  for (int i = 0; i < original_funcs.size(); ++i) {
+    ir::Expr func_body              = updated_bodys.at(i);
+    const ir::LoweredFunc& ori_func = original_funcs.at(i);
+    auto&& new_func                 = auto_schedule::UpdateFuncWithNewBody(target, ori_func, func_body);
+    results.emplace_back(new_func);
+  }
+
+  return results;
+}
+
+ir::Module BuildIRModule(const std::vector<ir::LoweredFunc>& lowered_funcs, const common::Target& target) {
+  ir::Module::Builder builder("test_bulder", target);
+  for (auto&& func : lowered_funcs) {
+    builder.AddFunction(func);
+  }
+  return builder.Build();
+}
+
+std::string GenSourceCode(const ir::Module& ir_module, const common::Target& target) {
+  std::unique_ptr<backends::CodeGenC> codegen;
+#ifdef CINN_WITH_CUDA
+  if (target == common::DefaultNVGPUTarget()) {
+    codegen = std::make_unique<backends::CodeGenCUDA_Dev>(target);
+  } else {
+    codegen = std::make_unique<backends::CodeGenCX86>(target, CodeGenCX86::Feature::AVX512);
+  }
+#else
+  codegen = std::make_unique<backends::CodeGenCX86>(target, CodeGenCX86::Feature::AVX512);
+#endif
+  codegen->SetInlineBuiltinCodes(false);
+  return codegen->Compile(ir_module, CodeGenC::OutputKind::CImpl);
+}
+
+std::vector<hlir::framework::Instruction> BuildExecution(const ir::Module& ir_module,
+                                                         const common::Target& target,
+                                                         hlir::framework::Scope* scope) {
+  auto backend_compier = backends::Compiler::Create(target);
+  backend_compier->Build(ir_module);
+
+  std::vector<hlir::framework::Instruction> results;
+  for (auto&& func : ir_module.functions()) {
+    std::vector<std::string> input_args;
+    std::vector<std::string> output_args;
+
+    for (auto&& arg : func->args) {
+      if (arg.is_input())
+        input_args.emplace_back(arg.name());
+      else
+        output_args.emplace_back(arg.name());
+    }
+    results.emplace_back(Instruction(target, scope, input_args, output_args, func->name));
+
+    auto func_ptr = reinterpret_cast<void (*)(void**, int32_t)>(backend_compier->Lookup(func->name));
+    results.back().SetLoweredFunc(reinterpret_cast<void*>(func_ptr));
+    results.back().Finalize();
+  }
+  return results;
+}
+
+}  // namespace tests
+}  // namespace cinn
diff --git a/tests/utils.h b/tests/utils.h
new file mode 100644
index 0000000000..fdb6902f32
--- /dev/null
+++ b/tests/utils.h
@@ -0,0 +1,52 @@
+// Copyright (c) 2023 CINN 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.
+
+#pragma once
+#include <string>
+#include <vector>
+
+#include "cinn/common/target.h"
+#include "cinn/frontend/optimize.h"
+#include "cinn/hlir/framework/graph.h"
+#include "cinn/hlir/framework/instruction.h"
+#include "cinn/hlir/framework/scope.h"
+#include "cinn/ir/ir_schedule.h"
+#include "cinn/ir/lowered_func.h"
+
+namespace cinn {
+namespace tests {
+
+std::shared_ptr<hlir::framework::Graph> OptimizeByPass(frontend::Program& program, const common::Target& target);
+
+std::vector<ir::LoweredFunc> LowerFusionGroup(std::shared_ptr<hlir::framework::Graph> graph,
+                                              std::shared_ptr<hlir::framework::Graph::Group> group,
+                                              const common::Target& target,
+                                              bool apply_manual_schedule = true);
+
+ir::IRSchedule MakeIRSchedule(const std::vector<ir::LoweredFunc>& lowered_funcs);
+
+std::vector<ir::LoweredFunc> OptimizeBySchedule(const ir::IRSchedule& schedule,
+                                                const std::vector<ir::LoweredFunc>& original_funcs,
+                                                const common::Target& target);
+
+ir::Module BuildIRModule(const std::vector<ir::LoweredFunc>& lowered_funcs, const common::Target& target);
+
+std::string GenSourceCode(const ir::Module& ir_module, const common::Target& target);
+
+std::vector<hlir::framework::Instruction> BuildExecution(const ir::Module& ir_module,
+                                                         const common::Target& target,
+                                                         hlir::framework::Scope* scope);
+
+}  // namespace tests
+}  // namespace cinn