diff --git a/litellm/llms/anthropic/chat/guardrail_translation/handler.py b/litellm/llms/anthropic/chat/guardrail_translation/handler.py
index cc0b260998f..06a1b92e1b0 100644
--- a/litellm/llms/anthropic/chat/guardrail_translation/handler.py
+++ b/litellm/llms/anthropic/chat/guardrail_translation/handler.py
@@ -13,7 +13,7 @@ Pattern Overview:
 """
 
 import asyncio
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, cast
+from typing import TYPE_CHECKING, Any, Coroutine, Dict, List, Optional, Tuple, cast
 
 from litellm._logging import verbose_proxy_logger
 from litellm.llms.base_llm.guardrail_translation.base_translation import BaseTranslation
@@ -49,7 +49,7 @@ class AnthropicMessagesHandler(BaseTranslation):
         if messages is None:
             return data
 
-        tasks = []
+        tasks: List[Coroutine[Any, Any, str]] = []
         task_mappings: List[Tuple[int, Optional[int]]] = []
         # Track (message_index, content_index) for each task
         # content_index is None for string content, int for list content
@@ -166,7 +166,7 @@ class AnthropicMessagesHandler(BaseTranslation):
             )
             return response
 
-        tasks = []
+        tasks: List[Coroutine[Any, Any, str]] = []
         task_mappings: List[Tuple[int, Optional[int]]] = []
         # Track (choice_index, content_index) for each task
 
diff --git a/litellm/llms/openai/chat/guardrail_translation/handler.py b/litellm/llms/openai/chat/guardrail_translation/handler.py
index 12d8cfad23a..ec25f491d8e 100644
--- a/litellm/llms/openai/chat/guardrail_translation/handler.py
+++ b/litellm/llms/openai/chat/guardrail_translation/handler.py
@@ -15,7 +15,7 @@ This pattern can be replicated for other message formats (e.g., Anthropic).
 """
 
 import asyncio
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, cast
+from typing import TYPE_CHECKING, Any, Coroutine, Dict, List, Optional, Tuple, cast
 
 import litellm
 from litellm._logging import verbose_proxy_logger
@@ -50,7 +50,7 @@ class OpenAIChatCompletionsHandler(BaseTranslation):
         if messages is None:
             return data
 
-        tasks = []
+        tasks: List[Coroutine[Any, Any, str]] = []
         task_mappings: List[Tuple[int, Optional[int]]] = []
         # Track (message_index, content_index) for each task
         # content_index is None for string content, int for list content
@@ -168,7 +168,7 @@ class OpenAIChatCompletionsHandler(BaseTranslation):
             )
             return response
 
-        tasks = []
+        tasks: List[Coroutine[Any, Any, str]] = []
         task_mappings: List[Tuple[int, Optional[int]]] = []
         # Track (choice_index, content_index) for each task
 
diff --git a/litellm/llms/openai/responses/guardrail_translation/handler.py b/litellm/llms/openai/responses/guardrail_translation/handler.py
index 40b1d73ddcb..fdac13176b1 100644
--- a/litellm/llms/openai/responses/guardrail_translation/handler.py
+++ b/litellm/llms/openai/responses/guardrail_translation/handler.py
@@ -29,7 +29,7 @@ Output: response.output is List[GenericResponseOutputItem] where each has:
 """
 
 import asyncio
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, Union, cast
+from typing import TYPE_CHECKING, Any, Coroutine, List, Optional, Tuple, Union, cast
 
 from litellm._logging import verbose_proxy_logger
 from litellm.llms.base_llm.guardrail_translation.base_translation import BaseTranslation
@@ -79,7 +79,7 @@ class OpenAIResponsesHandler(BaseTranslation):
         if not isinstance(input_data, list):
             return data
 
-        tasks = []
+        tasks: List[Coroutine[Any, Any, str]] = []
         task_mappings: List[Tuple[int, Optional[int]]] = []
         # Track (message_index, content_index) for each task
         # content_index is None for string content, int for list content
@@ -113,9 +113,9 @@ class OpenAIResponsesHandler(BaseTranslation):
 
     async def _extract_input_text_and_create_tasks(
         self,
-        message: Dict[str, Any],
+        message: Any,  # Can be Dict[str, Any] or ResponseInputParam
         msg_idx: int,
-        tasks: List,
+        tasks: List[Coroutine[Any, Any, str]],
         task_mappings: List[Tuple[int, Optional[int]]],
         guardrail_to_apply: "CustomGuardrail",
     ) -> None:
@@ -144,7 +144,7 @@ class OpenAIResponsesHandler(BaseTranslation):
 
     async def _apply_guardrail_responses_to_input(
         self,
-        messages: List[Dict[str, Any]],
+        messages: Any,  # Can be List[Dict[str, Any]] or ResponseInputParam
         responses: List[str],
         task_mappings: List[Tuple[int, Optional[int]]],
     ) -> None:
@@ -200,7 +200,7 @@ class OpenAIResponsesHandler(BaseTranslation):
             )
             return response
 
-        tasks = []
+        tasks: List[Coroutine[Any, Any, str]] = []
         task_mappings: List[Tuple[int, int]] = []
         # Track (output_item_index, content_index) for each task
 
diff --git a/litellm/proxy/guardrails/guardrail_hooks/grayswan/grayswan.py b/litellm/proxy/guardrails/guardrail_hooks/grayswan/grayswan.py
index 00e30d3e5ba..fb93dec47b7 100644
--- a/litellm/proxy/guardrails/guardrail_hooks/grayswan/grayswan.py
+++ b/litellm/proxy/guardrails/guardrail_hooks/grayswan/grayswan.py
@@ -216,7 +216,7 @@ class GraySwanGuardrail(CustomGuardrail):
 
         verbose_proxy_logger.debug("GraySwan Guardrail: post-call hook triggered")
 
-        response_dict = response.model_dump() if hasattr(response, "model_dump") else {}
+        response_dict = response.model_dump() if hasattr(response, "model_dump") else {}  # type: ignore[union-attr]
         response_messages = [
             msg if isinstance(msg, dict) else msg.model_dump()
             for choice in response_dict.get("choices", [])
diff --git a/litellm/proxy/guardrails/guardrail_hooks/pillar/pillar.py b/litellm/proxy/guardrails/guardrail_hooks/pillar/pillar.py
index 537b6038b53..1646e39b24d 100644
--- a/litellm/proxy/guardrails/guardrail_hooks/pillar/pillar.py
+++ b/litellm/proxy/guardrails/guardrail_hooks/pillar/pillar.py
@@ -284,7 +284,7 @@ class PillarGuardrail(CustomGuardrail):
         verbose_proxy_logger.debug("Pillar Guardrail: Post-call hook")
 
         # Extract response messages in the format Pillar expects
-        response_dict = response.model_dump() if hasattr(response, "model_dump") else {}
+        response_dict = response.model_dump() if hasattr(response, "model_dump") else {}  # type: ignore[union-attr]
         response_messages = [
             choice.get("message")
             for choice in response_dict.get("choices", [])
diff --git a/litellm/proxy/health_endpoints/_health_endpoints.py b/litellm/proxy/health_endpoints/_health_endpoints.py
index 216de57769e..c59e91aa1ef 100644
--- a/litellm/proxy/health_endpoints/_health_endpoints.py
+++ b/litellm/proxy/health_endpoints/_health_endpoints.py
@@ -926,6 +926,7 @@ async def test_model_connection(
             "batch",
             "rerank",
             "realtime",
+            "responses",
             "ocr",
         ]
     ] = fastapi.Body("chat", description="The mode to test the model with"),