fix - /moderation don't require a model

litellm/llms/openai.py

@@ -600,7 +600,7 @@ class OpenAIChatCompletion(BaseLLM):
         api_key: Optional[str] = None,
         api_base: Optional[str] = None,
         timeout: Union[float, httpx.Timeout] = httpx.Timeout(None),
-        max_retries: Optional[int] = None,
+        max_retries: Optional[int] = 2,
         organization: Optional[str] = None,
         client: Optional[Union[OpenAI, AsyncOpenAI]] = None,
     ):

litellm/main.py

@@ -3852,14 +3852,20 @@ def moderation(
 
 
 @client
-async def amoderation(input: str, model: str, api_key: Optional[str] = None, **kwargs):
+async def amoderation(
+    input: str, model: Optional[str] = None, api_key: Optional[str] = None, **kwargs
+):
     # only supports open ai for now
     api_key = (
         api_key or litellm.api_key or litellm.openai_key or get_secret("OPENAI_API_KEY")
     )
     openai_client = kwargs.get("client", None)
     if openai_client is None:
-        openai_client = openai.AsyncOpenAI(
+
+        # call helper to get OpenAI client
+        # _get_openai_client maintains in-memory caching logic for OpenAI clients
+        openai_client = openai_chat_completions._get_openai_client(
+            is_async=True,
             api_key=api_key,
         )
     response = await openai_client.moderations.create(input=input, model=model)

litellm/proxy/proxy_server.py

@@ -4947,7 +4947,7 @@ async def moderations(
         data["model"] = (
             general_settings.get("moderation_model", None)  # server default
             or user_model  # model name passed via cli args
-            or data["model"]  # default passed in http request
+            or data.get("model")  # default passed in http request
         )
         if user_model:
             data["model"] = user_model
@@ -4966,37 +4966,33 @@ async def moderations(
         if "api_key" in data:
             response = await litellm.amoderation(**data)
         elif (
-            llm_router is not None and data["model"] in router_model_names
+            llm_router is not None and data.get("model") in router_model_names
         ):  # model in router model list
             response = await llm_router.amoderation(**data)
         elif (
-            llm_router is not None and data["model"] in llm_router.deployment_names
+            llm_router is not None and data.get("model") in llm_router.deployment_names
         ):  # model in router deployments, calling a specific deployment on the router
             response = await llm_router.amoderation(**data, specific_deployment=True)
         elif (
             llm_router is not None
             and llm_router.model_group_alias is not None
-            and data["model"] in llm_router.model_group_alias
+            and data.get("model") in llm_router.model_group_alias
         ):  # model set in model_group_alias
             response = await llm_router.amoderation(
                 **data
             )  # ensure this goes the llm_router, router will do the correct alias mapping
         elif (
             llm_router is not None
-            and data["model"] not in router_model_names
+            and data.get("model") not in router_model_names
             and llm_router.default_deployment is not None
         ):  # model in router deployments, calling a specific deployment on the router
             response = await llm_router.amoderation(**data)
         elif user_model is not None:  # `litellm --model <your-model-name>`
             response = await litellm.amoderation(**data)
         else:
-            raise HTTPException(
+            # /moderations does not need a "model" passed
-                status_code=status.HTTP_400_BAD_REQUEST,
+            # see https://platform.openai.com/docs/api-reference/moderations
-                detail={
+            response = await litellm.amoderation(**data)
-                    "error": "moderations: Invalid model name passed in model="
-                    + data.get("model", "")
-                },
-            )
 
         ### ALERTING ###
         data["litellm_status"] = "success"  # used for alerting