diff --git a/litellm/__pycache__/utils.cpython-311.pyc b/litellm/__pycache__/utils.cpython-311.pyc
index c4098f31b..189cf63d1 100644
Binary files a/litellm/__pycache__/utils.cpython-311.pyc and b/litellm/__pycache__/utils.cpython-311.pyc differ
diff --git a/litellm/tests/test_streaming.py b/litellm/tests/test_streaming.py
index 7a20c6cb2..0ec8563ab 100644
--- a/litellm/tests/test_streaming.py
+++ b/litellm/tests/test_streaming.py
@@ -40,15 +40,13 @@ def test_completion_cohere_stream():
         # Add any assertions here to check the response
         for chunk in response:
             print(f"chunk: {chunk}")
-            if "content" in chunk["choices"][0]["delta"]:
-                complete_response += chunk["choices"][0]["delta"]["content"]
+            complete_response += chunk["choices"][0]["delta"]["content"]
         if complete_response == "": 
             raise Exception("Empty response received")
         print(f"completion_response: {complete_response}")
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
-
-
+        
 # test on baseten completion call
 # try:
 #     response = completion(
@@ -290,8 +288,6 @@ async def ai21_async_completion_call():
             complete_response += chunk["choices"][0]["delta"]["content"]
         if complete_response == "": 
             raise Exception("Empty response received")
-    except KeyError as e:
-        pass
     except:
         print(f"error occurred: {traceback.format_exc()}")
         pass
\ No newline at end of file
diff --git a/litellm/utils.py b/litellm/utils.py
index 2a3544abb..77cffe3ad 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -103,7 +103,7 @@ class Choices(OpenAIObject):
         self.message = message
 
 class StreamingChoices(OpenAIObject):
-    def __init__(self, finish_reason=None, index=0, delta: Optional[Union[Dict, Delta]]={}, **params):
+    def __init__(self, finish_reason=None, index=0, delta=Delta(), **params):
         super(StreamingChoices, self).__init__(**params)
         self.finish_reason = finish_reason
         self.index = index
@@ -2492,11 +2492,12 @@ class CustomStreamWrapper:
             model_response = ModelResponse(stream=True)
             model_response.choices[0].delta = completion_obj
             return model_response
+        except StopIteration:
+            raise StopIteration
         except Exception as e:
             model_response = ModelResponse(stream=True)
             model_response.choices[0].finish_reason = "stop"
             return model_response
-            # raise StopIteration
     
     async def __anext__(self):
         try: