diff --git a/litellm/llms/azure.py b/litellm/llms/azure.py
index 0371fc6c66..7a7a38b79d 100644
--- a/litellm/llms/azure.py
+++ b/litellm/llms/azure.py
@@ -111,14 +111,47 @@ class AzureChatCompletion(BaseLLM):
         exception_mapping_worked = False
         try:
 
-            if (model is None or messages is None) and client is None:
+            if model is None or messages is None:
                 raise AzureOpenAIError(status_code=422, message=f"Missing model or messages")
-        
-            data = {
-                "model": model,
-                "messages": messages, 
-                **optional_params
-            }
+            
+            max_retries = optional_params.pop("max_retries", 2)
+
+            ### CHECK IF CLOUDFLARE AI GATEWAY ###
+            ### if so - set the model as part of the base url 
+            if "gateway.ai.cloudflare.com" in api_base and client is None: 
+                ## build base url - assume api base includes resource name
+                if not api_base.endswith("/"): 
+                    api_base += "/"
+                api_base += f"{model}"
+                
+                azure_client_params = {
+                    "api_version": api_version,
+                    "base_url": f"{api_base}",
+                    "http_client": litellm.client_session,
+                    "max_retries": max_retries,
+                    "timeout": timeout
+                }
+                if api_key is not None:
+                    azure_client_params["api_key"] = api_key
+                elif azure_ad_token is not None:
+                    azure_client_params["azure_ad_token"] = azure_ad_token
+
+                if acompletion is True:
+                    client = AsyncAzureOpenAI(**azure_client_params)
+                else:
+                    client = AzureOpenAI(**azure_client_params)
+                
+                data = {
+                    "model": None,
+                    "messages": messages, 
+                    **optional_params
+                }
+            else: 
+                data = {
+                    "model": model, # type: ignore
+                    "messages": messages, 
+                    **optional_params
+                }
             ## LOGGING
             logging_obj.pre_call(
                 input=messages,
@@ -133,7 +166,7 @@ class AzureChatCompletion(BaseLLM):
                     "complete_input_dict": data,
                 },
             )
-            max_retries = data.pop("max_retries", 2)
+            
             if acompletion is True: 
                 if optional_params.get("stream", False):
                     return self.async_streaming(logging_obj=logging_obj, api_base=api_base, data=data, model=model, api_key=api_key, api_version=api_version, azure_ad_token=azure_ad_token, timeout=timeout, client=client)
diff --git a/litellm/main.py b/litellm/main.py
index fa5a8c868a..850d256723 100644
--- a/litellm/main.py
+++ b/litellm/main.py
@@ -494,31 +494,6 @@ def completion(
                 if k not in optional_params: # completion(top_k=3) > azure_config(top_k=3) <- allows for dynamic variables to be passed in
                     optional_params[k] = v
 
-            ### CHECK IF CLOUDFLARE AI GATEWAY ###
-            ### if so - set the model as part of the base url 
-            if "gateway.ai.cloudflare.com" in api_base and client is None: 
-                ## build base url - assume api base includes resource name
-                if not api_base.endswith("/"): 
-                    api_base += "/"
-                api_base += f"{model}"
-                
-                azure_client_params = {
-                    "api_version": api_version,
-                    "base_url": f"{api_base}",
-                    "http_client": litellm.client_session,
-                    "max_retries": max_retries,
-                    "timeout": timeout
-                }
-                if api_key is not None:
-                    azure_client_params["api_key"] = api_key
-                elif azure_ad_token is not None:
-                    azure_client_params["azure_ad_token"] = azure_ad_token
-                if acompletion is True:
-                    client = openai.AsyncAzureOpenAI(**azure_client_params)
-                else:
-                    client = openai.AzureOpenAI(**azure_client_params)
-                model = None
-
             ## COMPLETION CALL
             response = azure_chat_completions.completion(
                 model=model,
diff --git a/litellm/tests/test_completion.py b/litellm/tests/test_completion.py
index 8412d0f192..dcc7719907 100644
--- a/litellm/tests/test_completion.py
+++ b/litellm/tests/test_completion.py
@@ -1334,7 +1334,7 @@ def test_azure_cloudflare_api():
         response = completion(model="azure/gpt-turbo", messages=messages, base_url=os.getenv("CLOUDFLARE_AZURE_BASE_URL"), api_key=os.getenv("AZURE_FRANCE_API_KEY"))
         print(f"response: {response}")
     except Exception as e: 
-        print(e)
+        traceback.print_exc()
         pass
 
 test_azure_cloudflare_api()