fix(proxy): use error_message variable in raise_rate_limit_error

The raise_rate_limit_error method constructed an error_message variable
correctly but then used an f-string that directly interpolated
additional_details, causing "None" to appear in the error message when
no additional details were provided.

Fixes #18460

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

litellm/proxy/hooks/parallel_request_limiter.py

@@ -132,7 +132,7 @@ class _PROXY_MaxParallelRequestsHandler(CustomLogger):
             error_message = error_message + " " + additional_details
         raise HTTPException(
             status_code=429,
-            detail=f"Max parallel request limit reached {additional_details}",
+            detail=error_message,
             headers={"retry-after": str(self.time_to_next_minute())},
         )
 

tests/test_litellm/proxy/hooks/test_parallel_request_limiter.py

@@ -0,0 +1,61 @@
+"""
+Unit Tests for the max parallel request limiter for the proxy
+"""
+
+import pytest
+from fastapi import HTTPException
+from unittest.mock import MagicMock
+
+from litellm.caching.caching import DualCache
+from litellm.proxy.hooks.parallel_request_limiter import (
+    _PROXY_MaxParallelRequestsHandler,
+)
+
+
+class TestRaiseRateLimitError:
+    """Tests for the raise_rate_limit_error method"""
+
+    def test_raise_rate_limit_error_without_additional_details(self):
+        """
+        Test that error message does not contain 'None' when additional_details is not provided.
+
+        This is a regression test for issue #18460 where the error message would
+        show "Max parallel request limit reached None" instead of just
+        "Max parallel request limit reached".
+        """
+        mock_cache = MagicMock(spec=DualCache)
+        mock_internal_cache = MagicMock()
+        mock_internal_cache.dual_cache = mock_cache
+
+        handler = _PROXY_MaxParallelRequestsHandler(
+            internal_usage_cache=mock_internal_cache
+        )
+
+        with pytest.raises(HTTPException) as exc_info:
+            handler.raise_rate_limit_error()
+
+        assert exc_info.value.status_code == 429
+        assert exc_info.value.detail == "Max parallel request limit reached"
+        assert "None" not in exc_info.value.detail
+        assert "retry-after" in exc_info.value.headers
+
+    def test_raise_rate_limit_error_with_additional_details(self):
+        """
+        Test that error message includes additional_details when provided.
+        """
+        mock_cache = MagicMock(spec=DualCache)
+        mock_internal_cache = MagicMock()
+        mock_internal_cache.dual_cache = mock_cache
+
+        handler = _PROXY_MaxParallelRequestsHandler(
+            internal_usage_cache=mock_internal_cache
+        )
+
+        additional_info = "Hit limit for api_key. Current limits: max_parallel_requests: 5"
+
+        with pytest.raises(HTTPException) as exc_info:
+            handler.raise_rate_limit_error(additional_details=additional_info)
+
+        assert exc_info.value.status_code == 429
+        assert exc_info.value.detail == f"Max parallel request limit reached {additional_info}"
+        assert "retry-after" in exc_info.value.headers