Merge branch 'main' into litellm_guardrail_pipelines_ui

litellm/llms/vertex_ai/gemini/transformation.py

@@ -533,11 +533,12 @@ def _pop_and_merge_extra_body(data: RequestBody, optional_params: dict) -> None:
     """Pop extra_body from optional_params and shallow-merge into data, deep-merging dict values."""
     extra_body: Optional[dict] = optional_params.pop("extra_body", None)
     if extra_body is not None:
+        data_dict: dict = data  # type: ignore[assignment]
         for k, v in extra_body.items():
-            if k in data and isinstance(data[k], dict) and isinstance(v, dict):
-                data[k].update(v)
+            if k in data_dict and isinstance(data_dict[k], dict) and isinstance(v, dict):
+                data_dict[k].update(v)
             else:
-                data[k] = v
+                data_dict[k] = v
 
 
 def _transform_request_body(

litellm/proxy/_experimental/mcp_server/server.py

@@ -2029,7 +2029,7 @@ if MCP_AVAILABLE:
             # Inject masked debug headers when client sends x-litellm-mcp-debug: true
             _debug_headers = MCPDebug.maybe_build_debug_headers(
                 raw_headers=raw_headers,
-                scope=scope,
+                scope=dict(scope),
                 mcp_servers=mcp_servers,
                 mcp_auth_header=mcp_auth_header,
                 mcp_server_auth_headers=mcp_server_auth_headers,

litellm/proxy/pass_through_endpoints/pass_through_endpoints.py

@@ -1193,14 +1193,11 @@ def create_pass_through_route(
                 final_query_params.update(query_params)
             # When a caller (e.g. bedrock_proxy_route) supplies a pre-built
             # body, use it instead of the body parsed from the raw request.
+            final_custom_body: Optional[dict] = None
             if custom_body is not None:
                 final_custom_body = custom_body
-            else:
-                final_custom_body = (
-                    custom_body_data
-                    if isinstance(custom_body_data, dict) or custom_body_data is None
-                    else None
-                )
+            elif isinstance(custom_body_data, dict):
+                final_custom_body = custom_body_data
 
             return await pass_through_request(  # type: ignore
                 request=request,

litellm/proxy/vector_store_endpoints/endpoints.py

@@ -1,4 +1,4 @@
-from typing import Dict, Optional
+from typing import Any, Dict, Optional
 
 from fastapi import APIRouter, Depends, HTTPException, Request, Response
 
@@ -230,7 +230,7 @@ async def vector_store_create(
             )
         
         # Get managed vector stores hook
-        managed_vector_stores = proxy_logging_obj.get_proxy_hook("managed_vector_stores")
+        managed_vector_stores: Any = proxy_logging_obj.get_proxy_hook("managed_vector_stores")
         if managed_vector_stores is None:
             raise HTTPException(
                 status_code=500,

litellm/rag/ingestion/vertex_ai_ingestion.py

@@ -10,7 +10,7 @@ Based on: https://docs.cloud.google.com/vertex-ai/generative-ai/docs/model-refer
 from __future__ import annotations
 
 import json
-from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple, cast
+from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
 
 from litellm._logging import verbose_logger
 from litellm.llms.custom_httpx.http_handler import (

litellm/responses/litellm_completion_transformation/transformation.py

@@ -1500,7 +1500,7 @@ class LiteLLMCompletionResponsesConfig:
             previous_response_id=getattr(
                 chat_completion_response, "previous_response_id", None
             ),
-            reasoning=Reasoning(),
+            reasoning=dict(Reasoning()),
             status=LiteLLMCompletionResponsesConfig._map_chat_completion_finish_reason_to_responses_status(
                 finish_reason
             ),
@@ -1516,7 +1516,7 @@ class LiteLLMCompletionResponsesConfig:
         # Surface provider-specific fields (generic passthrough from any provider)
         provider_fields = responses_api_response._hidden_params.get("provider_specific_fields")
         if provider_fields:
-            responses_api_response.provider_specific_fields = provider_fields
+            setattr(responses_api_response, "provider_specific_fields", provider_fields)
 
         return responses_api_response
 

litellm/types/proxy/guardrails/guardrail_hooks/zscaler_ai_guard.py

@@ -106,6 +106,7 @@ class ZscalerAIGuardConfigModel(GuardrailConfigModel):
                     )
 
         # Check for configuration issues
+        assert api_base is not None  # always set via env default above
         is_resolve_policy = api_base.endswith("/resolve-and-execute-policy")
         is_execute_policy = api_base.endswith("/execute-policy") and not is_resolve_policy
 

model_prices_and_context_window.json

@@ -14835,7 +14835,9 @@
         "supports_tool_choice": true,
         "supports_url_context": true,
         "supports_vision": true,
-        "supports_web_search": true
+        "supports_web_search": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini-2.5-computer-use-preview-10-2025": {
         "input_cost_per_token": 1.25e-06,
@@ -16323,7 +16325,9 @@
         "source": "https://ai.google.dev/pricing",
         "supported_endpoints": [
             "/v1/audio/speech"
-        ]
+        ],
+        "tpm": 4000000,
+        "rpm": 10
     },
     "gemini/gemini-2.5-pro": {
         "cache_read_input_token_cost": 1.25e-07,
@@ -16821,7 +16825,9 @@
         "source": "https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models#foundation_models",
         "supports_function_calling": true,
         "supports_tool_choice": true,
-        "supports_vision": true
+        "supports_vision": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini/gemini-gemma-2-9b-it": {
         "input_cost_per_token": 3.5e-07,
@@ -16833,7 +16839,9 @@
         "source": "https://cloud.google.com/vertex-ai/generative-ai/docs/learn/models#foundation_models",
         "supports_function_calling": true,
         "supports_tool_choice": true,
-        "supports_vision": true
+        "supports_vision": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini/gemini-pro": {
         "input_cost_per_token": 3.5e-07,
@@ -36495,7 +36503,9 @@
             "text",
             "image"
         ],
-        "supports_vision": true
+        "supports_vision": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini/gemini-2.0-flash-lite-001": {
         "cache_read_input_token_cost": 1.875e-08,
@@ -36628,7 +36638,9 @@
             "audio"
         ],
         "supports_audio_input": true,
-        "supports_audio_output": true
+        "supports_audio_output": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini/gemini-2.5-flash-native-audio-preview-09-2025": {
         "input_cost_per_audio_token": 1e-06,
@@ -36652,7 +36664,9 @@
             "audio"
         ],
         "supports_audio_input": true,
-        "supports_audio_output": true
+        "supports_audio_output": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini/gemini-2.5-flash-native-audio-preview-12-2025": {
         "input_cost_per_audio_token": 1e-06,
@@ -36676,7 +36690,9 @@
             "audio"
         ],
         "supports_audio_input": true,
-        "supports_audio_output": true
+        "supports_audio_output": true,
+        "tpm": 250000,
+        "rpm": 10
     },
     "gemini-2.5-flash-preview-tts": {
         "input_cost_per_token": 3e-07,

ui/litellm-dashboard/e2e_tests/globalSetup.ts

@@ -8,7 +8,7 @@ async function globalSetup() {
   await page.goto("http://localhost:4000/ui/login");
   await page.getByPlaceholder("Enter your username").fill(users[Role.ProxyAdmin].email);
   await page.getByPlaceholder("Enter your password").fill(users[Role.ProxyAdmin].password);
-  const loginButton = page.getByRole("button", { name: "Login" });
+  const loginButton = page.getByRole("button", { name: "Login", exact: true });
   await loginButton.click();
   await page.waitForSelector("text=AI Gateway");
   await page.context().storageState({ path: "admin.storageState.json" });

ui/litellm-dashboard/e2e_tests/tests/login/login.spec.ts

@@ -6,7 +6,7 @@ test("user can log in", async ({ page }) => {
   await page.goto("http://localhost:4000/ui/login");
   await page.getByPlaceholder("Enter your username").fill(users[Role.ProxyAdmin].email);
   await page.getByPlaceholder("Enter your password").fill(users[Role.ProxyAdmin].password);
-  const loginButton = page.getByRole("button", { name: "Login" });
+  const loginButton = page.getByRole("button", { name: "Login", exact: true });
   await expect(loginButton).toBeEnabled();
   await loginButton.click();
   await expect(page.getByText("AI Gateway")).toBeVisible();