diff --git a/tests/test_litellm/llms/openai/test_gpt5_transformation.py b/tests/test_litellm/llms/openai/test_gpt5_transformation.py
index b136f8774be..f12e73aaeb1 100644
--- a/tests/test_litellm/llms/openai/test_gpt5_transformation.py
+++ b/tests/test_litellm/llms/openai/test_gpt5_transformation.py
@@ -350,7 +350,7 @@ def test_gpt5_normalizes_reasoning_effort_dict_from_optional_params(config: Open
 
 
 def test_gpt5_4_drops_reasoning_effort_when_tools_present(config: OpenAIConfig):
-    """gpt-5.4: function calls not supported with reasoning_effort != 'none'. Drop reasoning_effort."""
+    """gpt-5.4: reasoning_effort is KEPT when tools are present (routes to Responses API)."""
     tools = [{"type": "function", "function": {"name": "test", "description": "test"}}]
     params = config.map_openai_params(
         non_default_params={"reasoning_effort": "high", "tools": tools},
@@ -358,7 +358,8 @@ def test_gpt5_4_drops_reasoning_effort_when_tools_present(config: OpenAIConfig):
         model="gpt-5.4",
         drop_params=False,
     )
-    assert "reasoning_effort" not in params
+    # Changed: reasoning_effort is now KEPT for GPT-5.4+ with tools (routes to Responses API)
+    assert params["reasoning_effort"] == "high"
     assert params["tools"] == tools
 
 
diff --git a/tests/test_litellm/test_openai_gpt54_responses_auto_routing.py b/tests/test_litellm/test_openai_gpt54_responses_auto_routing.py
new file mode 100644
index 00000000000..af8cea4fcf0
--- /dev/null
+++ b/tests/test_litellm/test_openai_gpt54_responses_auto_routing.py
@@ -0,0 +1,328 @@
+"""
+Test automatic routing to OpenAI Responses API for GPT-5.4+ models with tools and reasoning
+"""
+import os
+import sys
+from unittest.mock import MagicMock, patch
+
+sys.path.insert(0, os.path.abspath("../.."))
+
+import pytest
+import litellm
+from litellm.main import responses_api_bridge_check
+
+
+class TestOpenAIGPT54ResponsesAutoRouting:
+    """Test that OpenAI GPT-5.4+ requests with tools and reasoning automatically route to Responses API"""
+
+    def test_gpt_54_with_tools_and_reasoning_routes_to_responses(self):
+        """Test that GPT-5.4 with tools and reasoning routes to Responses API"""
+        model = "gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                    "parameters": {
+                        "type": "object",
+                        "properties": {
+                            "location": {"type": "string"}
+                        }
+                    }
+                }
+            }
+        ]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") == "responses"
+        assert updated_model == model
+
+    def test_gpt_54_pro_with_tools_and_reasoning_routes_to_responses(self):
+        """Test that GPT-5.4-pro with tools and reasoning routes to Responses API"""
+        model = "gpt-5.4-pro"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "calculate",
+                    "description": "Perform calculation",
+                }
+            }
+        ]
+        reasoning_effort = "medium"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") == "responses"
+        assert updated_model == model
+
+    def test_gpt_54_with_tools_but_no_reasoning_does_not_route(self):
+        """Test that GPT-5.4 with tools but no reasoning does NOT route to Responses API"""
+        model = "gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                }
+            }
+        ]
+        reasoning_effort = None
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_54_with_reasoning_none_does_not_route(self):
+        """Test that GPT-5.4 with reasoning_effort='none' does NOT route to Responses API"""
+        model = "gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                }
+            }
+        ]
+        reasoning_effort = "none"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_54_with_reasoning_but_no_tools_does_not_route(self):
+        """Test that GPT-5.4 with reasoning but no tools does NOT route to Responses API"""
+        model = "gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = None
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_53_with_tools_and_reasoning_does_not_route(self):
+        """Test that GPT-5.3 (below 5.4) does NOT route to Responses API"""
+        model = "gpt-5.3"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                }
+            }
+        ]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_52_with_tools_and_reasoning_does_not_route(self):
+        """Test that GPT-5.2 (below 5.4) does NOT route to Responses API"""
+        model = "gpt-5.2"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                }
+            }
+        ]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_55_with_tools_and_reasoning_routes_to_responses(self):
+        """Test that GPT-5.5 (above 5.4) routes to Responses API"""
+        model = "gpt-5.5"
+        custom_llm_provider = "openai"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                }
+            }
+        ]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") == "responses"
+        assert updated_model == model
+
+    def test_azure_provider_with_gpt_54_does_not_route(self):
+        """Test that Azure provider does NOT auto-route (only OpenAI)"""
+        model = "gpt-5.4"
+        custom_llm_provider = "azure"
+        tools = [
+            {
+                "type": "function",
+                "function": {
+                    "name": "get_weather",
+                    "description": "Get the weather",
+                }
+            }
+        ]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_54_with_empty_tools_list_does_not_route(self):
+        """Test that GPT-5.4 with empty tools list does NOT route to Responses API"""
+        model = "gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = []
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") != "responses"
+        assert updated_model == model
+
+    def test_gpt_54_with_responses_prefix_still_routes(self):
+        """Test that responses/ prefix still works for GPT-5.4"""
+        model = "responses/gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = None
+        reasoning_effort = None
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") == "responses"
+        assert updated_model == "gpt-5.4"
+
+    def test_gpt_54_with_various_reasoning_levels(self):
+        """Test different reasoning_effort levels"""
+        model = "gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = [{"type": "function", "function": {"name": "test"}}]
+
+        for reasoning_level in ["low", "medium", "high", "xhigh"]:
+            model_info, updated_model = responses_api_bridge_check(
+                model=model,
+                custom_llm_provider=custom_llm_provider,
+                tools=tools,
+                reasoning_effort=reasoning_level,
+            )
+
+            assert model_info.get("mode") == "responses", f"Failed for reasoning_effort={reasoning_level}"
+            assert updated_model == model
+
+    def test_gpt_54_model_with_date_suffix(self):
+        """Test GPT-5.4 model with date suffix (e.g., gpt-5.4-2026-03-05)"""
+        model = "gpt-5.4-2026-03-05"
+        custom_llm_provider = "openai"
+        tools = [{"type": "function", "function": {"name": "test"}}]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") == "responses"
+        assert updated_model == model
+
+    def test_gpt_54_with_openai_prefix(self):
+        """Test GPT-5.4 with openai/ prefix"""
+        model = "openai/gpt-5.4"
+        custom_llm_provider = "openai"
+        tools = [{"type": "function", "function": {"name": "test"}}]
+        reasoning_effort = "high"
+
+        model_info, updated_model = responses_api_bridge_check(
+            model=model,
+            custom_llm_provider=custom_llm_provider,
+            tools=tools,
+            reasoning_effort=reasoning_effort,
+        )
+
+        assert model_info.get("mode") == "responses"
+        assert updated_model == model
+
+
+if __name__ == "__main__":
+    pytest.main([__file__, "-v"])