diff --git a/litellm/__init__.py b/litellm/__init__.py
index 5812e89bba9..670c19f4c13 100644
--- a/litellm/__init__.py
+++ b/litellm/__init__.py
@@ -531,7 +531,7 @@ from .llms.bedrock import (
     AmazonLlamaConfig,
 )
 from .llms.openai import OpenAIConfig, OpenAITextCompletionConfig
-from .llms.azure import AzureOpenAIConfig
+from .llms.azure import AzureOpenAIConfig, AzureOpenAIError
 from .main import *  # type: ignore
 from .integrations import *
 from .exceptions import (
diff --git a/litellm/tests/test_stream_chunk_builder.py b/litellm/tests/test_stream_chunk_builder.py
index f73265d6588..06ca04116d3 100644
--- a/litellm/tests/test_stream_chunk_builder.py
+++ b/litellm/tests/test_stream_chunk_builder.py
@@ -93,7 +93,7 @@ def test_stream_chunk_builder_litellm_function_call():
 
 def test_stream_chunk_builder_litellm_tool_call():
     try:
-        litellm.set_verbose = False
+        litellm.set_verbose = True
         response = litellm.completion(
             model="azure/gpt-4-nov-release",
             messages=messages,
@@ -101,6 +101,7 @@ def test_stream_chunk_builder_litellm_tool_call():
             stream=True,
             api_key="os.environ/AZURE_FRANCE_API_KEY",
             api_base="https://openai-france-1234.openai.azure.com",
+            api_version="2023-12-01-preview",
             complete_response=True,
         )
 
diff --git a/litellm/utils.py b/litellm/utils.py
index 5b6af9271d3..77e4de8e913 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -7147,6 +7147,16 @@ class CustomStreamWrapper:
                 else:
                     logprobs = None
 
+                if (
+                    hasattr(str_line.choices[0], "content_filter_result")
+                    and str_line.choices[0].content_filter_result is not None
+                ):
+                    error_message = json.dumps(
+                        str_line.choices[0].content_filter_result
+                    )
+                    raise litellm.AzureOpenAIError(
+                        status_code=400, message=error_message
+                    )
             return {
                 "text": text,
                 "is_finished": is_finished,
@@ -7695,7 +7705,6 @@ class CustomStreamWrapper:
                     chunk = self.completion_stream
                 else:
                     chunk = next(self.completion_stream)
-                print_verbose(f"value of chunk: {chunk} ")
                 if chunk is not None and chunk != b"":
                     print_verbose(f"PROCESSED CHUNK PRE CHUNK CREATOR: {chunk}")
                     response = self.chunk_creator(chunk=chunk)