Fix:test_aaaaazure_tenant_id_auth

litellm/llms/azure/azure.py

@@ -215,7 +215,7 @@ class AzureChatCompletion(BaseAzureLLM, BaseLLM):
 
             ### CHECK IF CLOUDFLARE AI GATEWAY ###
             ### if so - set the model as part of the base url
-            if "gateway.ai.cloudflare.com" in api_base:
+            if api_base is not None and "gateway.ai.cloudflare.com" in api_base:
                 client = self._init_azure_client_for_cloudflare_ai_gateway(
                     api_base=api_base,
                     model=model,
@@ -1338,7 +1338,7 @@ class AzureChatCompletion(BaseAzureLLM, BaseLLM):
         prompt: Optional[str] = None,
     ) -> dict:
         client_session = litellm.client_session or httpx.Client()
-        if "gateway.ai.cloudflare.com" in api_base:
+        if api_base is not None and "gateway.ai.cloudflare.com" in api_base:
             ## build base url - assume api base includes resource name
             if not api_base.endswith("/"):
                 api_base += "/"

litellm/llms/custom_httpx/http_handler.py

@@ -1168,8 +1168,10 @@ def get_async_httpx_client(
         return _cached_client
 
     if params is not None:
-        params["shared_session"] = shared_session
-        _new_client = AsyncHTTPHandler(**params)
+        # Filter out params that are only used for cache key, not for AsyncHTTPHandler.__init__
+        handler_params = {k: v for k, v in params.items() if k != "disable_aiohttp_transport"}
+        handler_params["shared_session"] = shared_session
+        _new_client = AsyncHTTPHandler(**handler_params)
     else:
         _new_client = AsyncHTTPHandler(
             timeout=httpx.Timeout(timeout=600.0, connect=5.0),
@@ -1215,7 +1217,9 @@ def _get_httpx_client(params: Optional[dict] = None) -> HTTPHandler:
         return _cached_client
 
     if params is not None:
-        _new_client = HTTPHandler(**params)
+        # Filter out params that are only used for cache key, not for HTTPHandler.__init__
+        handler_params = {k: v for k, v in params.items() if k != "disable_aiohttp_transport"}
+        _new_client = HTTPHandler(**handler_params)
     else:
         _new_client = HTTPHandler(timeout=httpx.Timeout(timeout=600.0, connect=5.0))
 

litellm/llms/openai/common_utils.py

@@ -212,7 +212,8 @@ class BaseOpenAILLM:
         try:
             # Get SSL config and include in params for proper cache key
             ssl_config = get_ssl_configuration()
-            params = {"ssl_verify": ssl_config} if ssl_config is not None else None
+            params = {"ssl_verify": ssl_config} if ssl_config is not None else {}
+            params["disable_aiohttp_transport"] = litellm.disable_aiohttp_transport
 
             # Get a cached AsyncHTTPHandler which manages the httpx.AsyncClient
             cached_handler = get_async_httpx_client(

tests/local_testing/test_azure_openai.py

@@ -41,6 +41,11 @@ async def test_aaaaazure_tenant_id_auth(respx_mock: MockRouter):
     PROD Test
     """
     litellm.disable_aiohttp_transport = True # since this uses respx, we need to set use_aiohttp_transport to False
+    
+    # Clear the HTTP client cache to ensure respx mocking works
+    # This is critical because respx only intercepts clients created AFTER mocking is active
+    if hasattr(litellm, 'in_memory_llm_clients_cache'):
+        litellm.in_memory_llm_clients_cache.flush_cache()
 
     router = Router(
         model_list=[