diff --git a/deepmd/train/run_options.py b/deepmd/train/run_options.py
index 8cd095323d..98ce9d3a9d 100644
--- a/deepmd/train/run_options.py
+++ b/deepmd/train/run_options.py
@@ -47,6 +47,7 @@
     f"source commit:        {GLOBAL_CONFIG['git_hash']}",
     f"source commit at:     {GLOBAL_CONFIG['git_date']}",
     f"build float prec:     {global_float_prec}",
+    f"build variant:        {GLOBAL_CONFIG['dp_variant']}",
     f"build with tf inc:    {GLOBAL_CONFIG['tf_include_dir']}",
     f"build with tf lib:    {GLOBAL_CONFIG['tf_libs'].replace(';', _sep)}"  # noqa
 )
diff --git a/source/CMakeLists.txt b/source/CMakeLists.txt
index 9f6351a911..9f276c39a0 100644
--- a/source/CMakeLists.txt
+++ b/source/CMakeLists.txt
@@ -53,6 +53,7 @@ message(STATUS "Supported model version: ${MODEL_VERSION}")
 if (USE_ROCM_TOOLKIT AND USE_CUDA_TOOLKIT)
   message (FATAL_ERROR "Devices that have both ROCM and CUDA are not currently supported")
 endif()
+set(DP_VARIANT "cpu")
 
 # define USE_CUDA_TOOLKIT
 if (USE_CUDA_TOOLKIT)
@@ -60,6 +61,7 @@ if (USE_CUDA_TOOLKIT)
   find_package(CUDA REQUIRED)
   add_definitions("-DGOOGLE_CUDA")
   message(STATUS "Found CUDA in ${CUDA_TOOLKIT_ROOT_DIR}, build nv GPU support")
+  set(DP_VARIANT "cuda")
 else()
   message(STATUS "Will not build nv GPU support")
 endif(USE_CUDA_TOOLKIT)
@@ -70,6 +72,7 @@ if (USE_ROCM_TOOLKIT)
   add_definitions("-DTENSORFLOW_USE_ROCM")
   add_compile_definitions(__HIP_PLATFORM_HCC__)
   message(STATUS "Found ROCM in ${ROCM_ROOT}, build AMD GPU support")
+  set(DP_VARIANT "rocm")
 else()
   message(STATUS "Will not build AMD GPU support")  
 endif (USE_ROCM_TOOLKIT)
diff --git a/source/api_cc/src/DeepPot.cc b/source/api_cc/src/DeepPot.cc
index 2b84b2405e..0dda00cb27 100644
--- a/source/api_cc/src/DeepPot.cc
+++ b/source/api_cc/src/DeepPot.cc
@@ -248,6 +248,13 @@ print_summary(const std::string &pre) const
   std::cout << pre << "source commit at:   " + global_git_date << std::endl;
   std::cout << pre << "surpport model ver.:" + global_model_version << std::endl;
   std::cout << pre << "build float prec:   " + global_float_prec << std::endl;
+#if defined(GOOGLE_CUDA)
+  std::cout << pre << "build variant:      cuda" << std::endl;
+#elif defined(TENSORFLOW_USE_ROCM)
+  std::cout << pre << "build variant:      rocm" << std::endl;
+#else
+  std::cout << pre << "build variant:      cpu" << std::endl;
+#endif
   std::cout << pre << "build with tf inc:  " + global_tf_include_dir << std::endl;
   std::cout << pre << "build with tf lib:  " + global_tf_lib << std::endl;
   std::cout << pre << "set tf intra_op_parallelism_threads: " <<  num_intra_nthreads << std::endl;
diff --git a/source/config/run_config.ini b/source/config/run_config.ini
index 87f54ef169..3f0a7a33a8 100644
--- a/source/config/run_config.ini
+++ b/source/config/run_config.ini
@@ -9,3 +9,4 @@ TF_LIBS = @TensorFlow_LIBRARY@
 TF_VERSION = @TENSORFLOW_VERSION@
 TF_CXX11_ABI_FLAG = @OP_CXX_ABI@
 MODEL_VERSION=@MODEL_VERSION@
+DP_VARIANT=@DP_VARIANT@