diff --git a/litellm/tests/test_completion.py b/litellm/tests/test_completion.py
index 51cad297a19..b6a3303c330 100644
--- a/litellm/tests/test_completion.py
+++ b/litellm/tests/test_completion.py
@@ -32,6 +32,7 @@ def test_completion_custom_provider_model_name():
         )
         # Add any assertions here to check the response
         print(response)
+        print(response['choices'][0]['finish_reason'])
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
 
@@ -107,8 +108,10 @@ def test_completion_claude_stream():
         # Add any assertions here to check the response
         for chunk in response:
             print(chunk["choices"][0]["delta"])  # same as openai format
+            print(chunk["choices"][0]["finish_reason"])
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
+# test_completion_claude_stream()
 
 
 
@@ -173,8 +176,10 @@ def test_completion_cohere_stream():
         # Add any assertions here to check the response
         for chunk in response:
             print(chunk["choices"][0]["delta"])  # same as openai format
+            print(chunk["choices"][0]["finish_reason"])
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
+# test_completion_cohere_stream()
 
 
 def test_completion_openai():
@@ -293,8 +298,12 @@ def test_completion_openai_with_stream():
         )
         # Add any assertions here to check the response
         print(response)
+        for chunk in response:
+            print(chunk)
+            print(chunk["choices"][0]["finish_reason"])
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
+# test_completion_openai_with_stream()
 
 
 def test_completion_openai_with_functions():
@@ -317,12 +326,16 @@ def test_completion_openai_with_functions():
     ]
     try:
         response = completion(
-            model="gpt-3.5-turbo", messages=messages, functions=function1
+            model="gpt-3.5-turbo", messages=messages, functions=function1, stream=True
         )
         # Add any assertions here to check the response
         print(response)
+        for chunk in response:
+            print(chunk)
+            print(chunk["choices"][0]["finish_reason"])
     except Exception as e:
         pytest.fail(f"Error occurred: {e}")
+# test_completion_openai_with_functions()
 
 
 def test_completion_azure():
diff --git a/litellm/utils.py b/litellm/utils.py
index 507e130f554..34c618b178a 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -2198,12 +2198,22 @@ class CustomStreamWrapper:
                 completion_obj["content"] = self.handle_openai_text_completion_chunk(chunk)
             else: # openai chat/azure models
                 chunk = next(self.completion_stream)
-                completion_obj["content"] = self.handle_openai_chat_completion_chunk(chunk)
+                return chunk # open ai returns finish_reason, we should just return the openai chunk
+            
+                #completion_obj["content"] = self.handle_openai_chat_completion_chunk(chunk)
             
             # LOGGING
             threading.Thread(target=self.logging_obj.success_handler, args=(completion_obj,)).start()
             # return this for all models
-            return {"choices": [{"delta": completion_obj}]}
+            return {
+                "choices": 
+                    [
+                        {
+                            "delta": completion_obj,
+                            "finish_reason": "stop"
+                        },
+                    ]
+                }
         except Exception as e:
             raise StopIteration
     
diff --git a/pyproject.toml b/pyproject.toml
index 8f8b312b2c5..0907ff1acb7 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "litellm"
-version = "0.1.580"
+version = "0.1.582"
 description = "Library to easily interface with LLM API providers"
 authors = ["BerriAI"]
 license = "MIT License"