Fix code qa and mypy lint issues

litellm/integrations/websearch_interception/handler.py

@@ -28,6 +28,7 @@ from litellm.integrations.websearch_interception.transformation import (
 from litellm.types.integrations.websearch_interception import (
     WebSearchInterceptionConfig,
 )
+from litellm.types.llms.openai import AllMessageValues
 from litellm.types.utils import LlmProviders
 from litellm.utils import ProviderConfigManager
 
@@ -138,7 +139,7 @@ class WebSearchInterceptionLogger(CustomLogger):
             get_last_user_message,
         )
 
-        query = get_last_user_message(messages)
+        query = get_last_user_message(cast(List[AllMessageValues], messages))
         if not query:
             return None
 

litellm/llms/anthropic/experimental_pass_through/messages/handler.py

@@ -8,7 +8,7 @@
 import asyncio
 import contextvars
 from functools import partial
-from typing import Any, AsyncIterator, Coroutine, Dict, List, Optional, Union
+from typing import Any, AsyncIterator, Coroutine, Dict, List, Optional, Union, cast
 
 import litellm
 from litellm.litellm_core_utils.litellm_logging import Logging as LiteLLMLoggingObj
@@ -151,13 +151,14 @@ async def _try_websearch_short_circuit(
             custom_llm_provider=custom_llm_provider,
         )
         if response is not None:
+            anthropic_response = cast(AnthropicMessagesResponse, response)
             if stream:
                 from litellm.llms.anthropic.experimental_pass_through.messages.fake_stream_iterator import (
                     FakeAnthropicMessagesStreamIterator,
                 )
 
-                return FakeAnthropicMessagesStreamIterator(response)
-            return response
+                return FakeAnthropicMessagesStreamIterator(anthropic_response)
+            return anthropic_response
 
     return None
 

litellm/llms/openai/chat/gpt_transformation.py

@@ -7,6 +7,7 @@ from typing import (
     Any,
     AsyncIterator,
     Coroutine,
+    Dict,
     Iterator,
     List,
     Literal,
@@ -805,7 +806,7 @@ class OpenAIChatCompletionStreamingHandler(BaseModelResponseIterator):
             choices = chunk.get("choices", [])
             choices = self._map_reasoning_to_reasoning_content(choices)
 
-            kwargs = {
+            kwargs: Dict[str, Any] = {
                 "id": chunk.get("id"),
                 "object": "chat.completion.chunk",
                 "created": chunk.get("created"),

litellm/proxy/auth/auth_utils.py

@@ -560,7 +560,8 @@ def _get_deployment_default_limit(model_name: str, field: str) -> Optional[int]:
         raw = deployment.get("litellm_params", {}).get(field)
         if raw is not None:
             try:
-                limits.append(int(raw))
+                if isinstance(raw, (int, float, str, bytes, bytearray)):
+                    limits.append(int(raw))
             except (ValueError, TypeError):
                 pass
     return min(limits) if limits else None

litellm/proxy/hooks/parallel_request_limiter.py

@@ -478,6 +478,15 @@ class _PROXY_MaxParallelRequestsHandler(CustomLogger):
                 kwargs["litellm_params"]["metadata"].get("user_api_key_metadata", {})
                 or {}
             )
+            user_api_key_team_metadata = kwargs["litellm_params"]["metadata"].get(
+                "user_api_key_team_metadata", None
+            )
+            user_api_key_dict = UserAPIKeyAuth(
+                api_key=user_api_key,
+                metadata=user_api_key_metadata,
+                model_max_budget=user_api_key_model_max_budget,
+                team_metadata=user_api_key_team_metadata,
+            )
 
             # ------------
             # Setup values