(feat) add moderation on router

litellm/router.py

@@ -599,6 +599,98 @@ class Router:
                 self.fail_calls[model_name] += 1
             raise e
 
+    async def amoderation(self, model: str, input: str, **kwargs):
+        try:
+            kwargs["model"] = model
+            kwargs["input"] = input
+            kwargs["original_function"] = self._amoderation
+            kwargs["num_retries"] = kwargs.get("num_retries", self.num_retries)
+            timeout = kwargs.get("request_timeout", self.timeout)
+            kwargs.setdefault("metadata", {}).update({"model_group": model})
+
+            response = await self.async_function_with_fallbacks(**kwargs)
+
+            return response
+        except Exception as e:
+            raise e
+
+    async def _amoderation(self, model: str, input: str, **kwargs):
+        model_name = None
+        try:
+            verbose_router_logger.debug(
+                f"Inside _moderation()- model: {model}; kwargs: {kwargs}"
+            )
+            deployment = self.get_available_deployment(
+                model=model,
+                input=input,
+                specific_deployment=kwargs.pop("specific_deployment", None),
+            )
+            kwargs.setdefault("metadata", {}).update(
+                {
+                    "deployment": deployment["litellm_params"]["model"],
+                    "model_info": deployment.get("model_info", {}),
+                }
+            )
+            kwargs["model_info"] = deployment.get("model_info", {})
+            data = deployment["litellm_params"].copy()
+            model_name = data["model"]
+            for k, v in self.default_litellm_params.items():
+                if (
+                    k not in kwargs and v is not None
+                ):  # prioritize model-specific params > default router params
+                    kwargs[k] = v
+                elif k == "metadata":
+                    kwargs[k].update(v)
+
+            potential_model_client = self._get_client(
+                deployment=deployment, kwargs=kwargs, client_type="async"
+            )
+            # check if provided keys == client keys #
+            dynamic_api_key = kwargs.get("api_key", None)
+            if (
+                dynamic_api_key is not None
+                and potential_model_client is not None
+                and dynamic_api_key != potential_model_client.api_key
+            ):
+                model_client = None
+            else:
+                model_client = potential_model_client
+            self.total_calls[model_name] += 1
+
+            timeout = (
+                data.get(
+                    "timeout", None
+                )  # timeout set on litellm_params for this deployment
+                or self.timeout  # timeout set on router
+                or kwargs.get(
+                    "timeout", None
+                )  # this uses default_litellm_params when nothing is set
+            )
+
+            response = await litellm.amoderation(
+                **{
+                    **data,
+                    "input": input,
+                    "caching": self.cache_responses,
+                    "client": model_client,
+                    "timeout": timeout,
+                    **kwargs,
+                }
+            )
+
+            self.success_calls[model_name] += 1
+            verbose_router_logger.info(
+                f"litellm.amoderation(model={model_name})\033[32m 200 OK\033[0m"
+            )
+            return response
+        except Exception as e:
+            verbose_router_logger.info(
+                f"litellm.amoderation(model={model_name})\033[31m Exception {str(e)}\033[0m"
+            )
+            if model_name is not None:
+                self.fail_calls[model_name] += 1
+            raise e
+
     def text_completion(
         self,
         model: str,