diff --git a/PaddleNLP/PaddleTextGEN/seq2seq/args.py b/PaddleNLP/PaddleTextGEN/seq2seq/args.py
index ee056e33597651f9e166e4d6399c89bfc36598f7..eba46cb3beed3b63ec73a9c58554a05993b6a08c 100644
--- a/PaddleNLP/PaddleTextGEN/seq2seq/args.py
+++ b/PaddleNLP/PaddleTextGEN/seq2seq/args.py
@@ -122,6 +122,16 @@ def parse_args():
 
     parser.add_argument(
         "--profile", action='store_true', help="Whether enable the profile.")
-
+    # NOTE: profiler args, used for benchmark
+    parser.add_argument(
+        "--is_profiler",
+        type=int,
+        default=0,
+        help="The switch of profiler tools. (used for benchmark)")
+    parser.add_argument(
+        "--profiler_path",
+        type=str,
+        default='./',
+        help="the profiler output file path. (used for benchmark)")
     args = parser.parse_args()
     return args
diff --git a/PaddleNLP/PaddleTextGEN/seq2seq/train.py b/PaddleNLP/PaddleTextGEN/seq2seq/train.py
index 51d4d29eac141f35676fe92ef2713c63f86d4aae..cbf6aa4f39af5322bc58dba7a60251eca83d02ff 100644
--- a/PaddleNLP/PaddleTextGEN/seq2seq/train.py
+++ b/PaddleNLP/PaddleTextGEN/seq2seq/train.py
@@ -27,6 +27,7 @@ import contextlib
 
 import paddle
 import paddle.fluid as fluid
+from paddle.fluid import profiler
 import paddle.fluid.framework as framework
 import paddle.fluid.profiler as profiler
 from paddle.fluid.executor import Executor
@@ -213,6 +214,13 @@ def main():
                     ce_ppl.append(np.exp(total_loss / word_count))
                     total_loss = 0.0
                     word_count = 0.0
+                
+                # profiler tools
+                if args.is_profiler and epoch_id == 0 and batch_id == 100:
+                    profiler.start_profiler("All")
+                elif args.is_profiler and epoch_id == 0 and batch_id == 105:
+                    profiler.stop_profiler("total", args.profiler_path)
+                    return
 
             end_time = time.time()
             epoch_time = end_time - start_time