diff --git a/slm/model_zoo/gpt-3/ppfleetx/core/engine/inference_engine.py b/slm/model_zoo/gpt-3/ppfleetx/core/engine/inference_engine.py
index 58e323d910cd..c9014bb595f5 100644
--- a/slm/model_zoo/gpt-3/ppfleetx/core/engine/inference_engine.py
+++ b/slm/model_zoo/gpt-3/ppfleetx/core/engine/inference_engine.py
@@ -18,6 +18,12 @@
 import numpy as np
 import paddle
 import paddle.distributed.fleet as fleet
+
+from paddlenlp.utils.env import (
+    PADDLE_INFERENCE_MODEL_SUFFIX,
+    PADDLE_INFERENCE_WEIGHTS_SUFFIX,
+)
+
 try:
     from ppfleetx_ops import topp_sampling
 except Exception as e:
@@ -153,9 +159,9 @@ def _check_model(self):
         model_files = []
         param_files = []
         for fname in os.listdir(rank_path):
-            if os.path.splitext(fname)[1] == ".pdmodel":
+            if os.path.splitext(fname)[1] == f"{PADDLE_INFERENCE_MODEL_SUFFIX}":
                 model_files.append(fname)
-            if os.path.splitext(fname)[1] == ".pdiparams":
+            if os.path.splitext(fname)[1] == f"{PADDLE_INFERENCE_WEIGHTS_SUFFIX}":
                 param_files.append(fname)
 
         def _check_and_get_file(files, tag):
@@ -184,8 +190,8 @@ def _generate_comm_init_config(self, rank, nranks):
 
     def _init_predictor(self):
         if self.auto:
-            self.model_file = os.path.join(self.model_dir, "auto_dist{}.pdmodel".format(self.rank))
-            self.param_file = os.path.join(self.model_dir, "auto_dist{}.pdiparams".format(self.rank))
+            self.model_file = os.path.join(self.model_dir, f"auto_dist{self.rank}{PADDLE_INFERENCE_MODEL_SUFFIX}")
+            self.param_file = os.path.join(self.model_dir, f"auto_dist{self.rank}{PADDLE_INFERENCE_WEIGHTS_SUFFIX}")
         config = paddle.inference.Config(self.model_file, self.param_file)
 
         config.enable_memory_optim()