From d77d525ce770558a5ef8242f68fe5e87fba359c8 Mon Sep 17 00:00:00 2001
From: ryan-crabbe <128659760+ryan-crabbe@users.noreply.github.com>
Date: Sat, 7 Feb 2026 09:18:57 -0800
Subject: [PATCH] perf: add LRU cache to normalize_request_route (#19812)

Add @lru_cache(maxsize=256) to eliminate redundant regex work for
repeated routes. Reduces time from 1.04s to ~0s for 6,006 calls.
---
 litellm/proxy/auth/auth_utils.py | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/litellm/proxy/auth/auth_utils.py b/litellm/proxy/auth/auth_utils.py
index 2bd84a1d98..fbfc88228c 100644
--- a/litellm/proxy/auth/auth_utils.py
+++ b/litellm/proxy/auth/auth_utils.py
@@ -1,6 +1,7 @@
 import os
 import re
 import sys
+from functools import lru_cache
 from typing import Any, List, Optional, Tuple
 
 from fastapi import HTTPException, Request, status
@@ -311,10 +312,11 @@ def get_request_route(request: Request) -> str:
         return request.url.path
 
 
+@lru_cache(maxsize=256)
 def normalize_request_route(route: str) -> str:
     """
     Normalize request routes by replacing dynamic path parameters with placeholders.
-    
+
     This prevents high cardinality in Prometheus metrics by collapsing routes like:
     - /v1/responses/1234567890 -> /v1/responses/{response_id}
     - /v1/threads/thread_123 -> /v1/threads/{thread_id}