fix(logging): keep proxy-side rate limits as expected client errors

ProxyRateLimitError derives from HTTPException but carries an llm_provider,
so it read as provider-originated and regained its traceback. Any
HTTPException is a proxy rejection regardless of llm_provider.

litellm/litellm_core_utils/core_helpers.py

@@ -58,7 +58,17 @@ def safe_divide(
     return numerator / denominator
 
 
+def _is_proxy_rejection(exception: BaseException) -> bool:
+    try:
+        from starlette.exceptions import HTTPException
+    except ImportError:
+        return False
+    return isinstance(exception, HTTPException)
+
+
 def _is_provider_originated(exception: BaseException) -> bool:
+    if _is_proxy_rejection(exception):
+        return False
     if getattr(exception, "llm_provider", None):
         return True
     from litellm.llms.base_llm.chat.transformation import BaseLLMException
@@ -74,6 +84,8 @@ def is_expected_client_error(exception: BaseException | None) -> bool:
     client error and keeps its traceback: a mapped litellm exception carries
     ``llm_provider``, and the raw ``BaseLLMException`` that provider handlers
     raise before mapping (the /v1/messages route surfaces it as-is) is one too.
+    The proxy's own limiters raise ``HTTPException`` subclasses that also carry
+    an ``llm_provider``, so any ``HTTPException`` stays a proxy rejection.
 
     ProxyException stores the status on .code (as a str), HTTPException and
     litellm exceptions on .status_code.

tests/test_litellm/litellm_core_utils/test_core_helpers.py

@@ -286,6 +286,7 @@ class TestIsExpectedClientError:
         from litellm.exceptions import AuthenticationError, RateLimitError
         from litellm.litellm_core_utils.core_helpers import is_expected_client_error
         from litellm.llms.anthropic.common_utils import AnthropicError
+        from litellm.proxy.common_utils.proxy_rate_limit_error import ProxyRateLimitError
 
         provider_auth_failure = AuthenticationError(
             message="AnthropicException - API key is invalid.", llm_provider="anthropic", model="claude-haiku-4-5"
@@ -298,6 +299,12 @@ class TestIsExpectedClientError:
         unmapped_provider_failure = AnthropicError(status_code=401, message='{"type":"authentication_error"}')
         assert is_expected_client_error(unmapped_provider_failure) is False
 
+        proxy_rate_limit = ProxyRateLimitError(
+            detail={"error": "Max parallel requests reached"}, model="claude-haiku-4-5", llm_provider="anthropic"
+        )
+        assert proxy_rate_limit.llm_provider == "anthropic"
+        assert is_expected_client_error(proxy_rate_limit) is True
+
         class RouterRejection(Exception):
             def __init__(self):
                 self.status_code = 429