diff --git a/bentoml/service/__init__.py b/bentoml/service/__init__.py
index 362e22e8558..244cc3a0b1c 100644
--- a/bentoml/service/__init__.py
+++ b/bentoml/service/__init__.py
@@ -66,14 +66,14 @@ def validate_inference_api_name(api_name: str):
         )
 
 
-def validate_inference_api_route(api_route: str):
-    if re.findall(r"[?#]+|^(//)|^:", api_route):
+def validate_inference_api_route(route: str):
+    if re.findall(r"[?#]+|^(//)|^:", route):
         raise InvalidArgument(
-            "The path {} contains illegal url characters".format(api_route)
+            "The path {} contains illegal url characters".format(route)
         )
-    if api_route in BENTOML_RESERVED_API_NAMES:
+    if route in BENTOML_RESERVED_API_NAMES:
         raise InvalidArgument(
-            "Reserved API route: '{}' is reserved for infra endpoints".format(api_route)
+            "Reserved API route: '{}' is reserved for infra endpoints".format(route)
         )
 
 
@@ -473,7 +473,7 @@ def _config_inference_apis(self):
         ):
             if hasattr(function, "_is_api"):
                 api_name = getattr(function, "_api_name")
-                api_route = getattr(function, "_api_route", None)
+                route = getattr(function, "_api_route", None)
                 api_doc = getattr(function, "_api_doc")
                 input_adapter = getattr(function, "_input_adapter")
                 output_adapter = getattr(function, "_output_adapter")
@@ -495,7 +495,7 @@ def _config_inference_apis(self):
                         mb_max_latency=mb_max_latency,
                         mb_max_batch_size=mb_max_batch_size,
                         batch=batch,
-                        api_route=api_route,
+                        route=route,
                     )
                 )
 
diff --git a/bentoml/service/inference_api.py b/bentoml/service/inference_api.py
index 6a11653f384..fa6f244ca0d 100644
--- a/bentoml/service/inference_api.py
+++ b/bentoml/service/inference_api.py
@@ -50,7 +50,7 @@ def __init__(
         mb_max_latency=10000,
         mb_max_batch_size=1000,
         batch=False,
-        api_route=None,
+        route=None,
     ):
         """
         :param service: ref to service containing this API
@@ -82,7 +82,7 @@ def __init__(
         self.mb_max_latency = mb_max_latency
         self.mb_max_batch_size = mb_max_batch_size
         self.batch = batch
-        self.route = name if api_route is None else api_route
+        self.route = name if route is None else route
 
         if not self.input_adapter.BATCH_MODE_SUPPORTED and batch:
             raise BentoMLConfigException(
diff --git a/tests/server/test_model_api_server.py b/tests/server/test_model_api_server.py
index ae3e05a7155..b01c6dc44b7 100644
--- a/tests/server/test_model_api_server.py
+++ b/tests/server/test_model_api_server.py
@@ -34,7 +34,7 @@ def test_api_function_route(bento_service, img_file):
 
     response = test_client.post("/v1!@$%^&*()_-+=[]\\|;:,./predict", data='{"a": 1}',)
     assert 200 == response.status_code
-    assert '{"a": 1}' == str(response.data)
+    assert '{"a": 1}' == response.data.decode()
 
     assert "predict_dataframe" in index_list
     data = [{"col1": 10}, {"col1": 20}]