diff --git a/litellm/constants.py b/litellm/constants.py
index 05ccfb5993..5ab56bcaa2 100644
--- a/litellm/constants.py
+++ b/litellm/constants.py
@@ -100,8 +100,10 @@ RUNWAYML_POLLING_TIMEOUT = int(
 ########## Networking constants ##############################################################
 _DEFAULT_TTL_FOR_HTTPX_CLIENTS = 3600  # 1 hour, re-use the same httpx client for 1 hour
 
-# Aiohttp connection pooling constants
-AIOHTTP_CONNECTOR_LIMIT = int(os.getenv("AIOHTTP_CONNECTOR_LIMIT", 0))
+# Aiohttp connection pooling - prevents memory leaks from unbounded connection growth
+# Set to 0 for unlimited (not recommended for production)
+AIOHTTP_CONNECTOR_LIMIT = int(os.getenv("AIOHTTP_CONNECTOR_LIMIT", 300))
+AIOHTTP_CONNECTOR_LIMIT_PER_HOST = int(os.getenv("AIOHTTP_CONNECTOR_LIMIT_PER_HOST", 50))
 AIOHTTP_KEEPALIVE_TIMEOUT = int(os.getenv("AIOHTTP_KEEPALIVE_TIMEOUT", 120))
 AIOHTTP_TTL_DNS_CACHE = int(os.getenv("AIOHTTP_TTL_DNS_CACHE", 300))
 # enable_cleanup_closed is only needed for Python versions with the SSL leak bug
diff --git a/litellm/llms/custom_httpx/http_handler.py b/litellm/llms/custom_httpx/http_handler.py
index b06e8463ab..d9a9d4f9dc 100644
--- a/litellm/llms/custom_httpx/http_handler.py
+++ b/litellm/llms/custom_httpx/http_handler.py
@@ -16,6 +16,7 @@ from litellm._logging import verbose_logger
 from litellm.constants import (
     _DEFAULT_TTL_FOR_HTTPX_CLIENTS,
     AIOHTTP_CONNECTOR_LIMIT,
+    AIOHTTP_CONNECTOR_LIMIT_PER_HOST,
     AIOHTTP_KEEPALIVE_TIMEOUT,
     AIOHTTP_NEEDS_CLEANUP_CLOSED,
     AIOHTTP_TTL_DNS_CACHE,
@@ -793,15 +794,20 @@ class AsyncHTTPHandler:
         verbose_logger.debug(
             "NEW SESSION: Creating new ClientSession (no shared session provided)"
         )
+        transport_connector_kwargs = {
+            "keepalive_timeout": AIOHTTP_KEEPALIVE_TIMEOUT,
+            "ttl_dns_cache": AIOHTTP_TTL_DNS_CACHE,
+            "enable_cleanup_closed": True,
+            **connector_kwargs,
+        }
+        if AIOHTTP_CONNECTOR_LIMIT > 0:
+            transport_connector_kwargs["limit"] = AIOHTTP_CONNECTOR_LIMIT
+        if AIOHTTP_CONNECTOR_LIMIT_PER_HOST > 0:
+            transport_connector_kwargs["limit_per_host"] = AIOHTTP_CONNECTOR_LIMIT_PER_HOST
+        
         return LiteLLMAiohttpTransport(
             client=lambda: ClientSession(
-                connector=TCPConnector(
-                    limit=AIOHTTP_CONNECTOR_LIMIT,
-                    keepalive_timeout=AIOHTTP_KEEPALIVE_TIMEOUT,
-                    ttl_dns_cache=AIOHTTP_TTL_DNS_CACHE,
-                    enable_cleanup_closed=AIOHTTP_NEEDS_CLEANUP_CLOSED,
-                    **connector_kwargs,
-                ),
+                connector=TCPConnector(**transport_connector_kwargs),
                 trust_env=trust_env,
             ),
         )
diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index a1e01caddc..f2c7c361c4 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -31,6 +31,7 @@ from typing import (
 from litellm._uuid import uuid
 from litellm.constants import (
     AIOHTTP_CONNECTOR_LIMIT,
+    AIOHTTP_CONNECTOR_LIMIT_PER_HOST,
     AIOHTTP_KEEPALIVE_TIMEOUT,
     AIOHTTP_NEEDS_CLEANUP_CLOSED,
     AIOHTTP_TTL_DNS_CACHE,
@@ -627,21 +628,26 @@ async def proxy_shutdown_event():
 
 
 async def _initialize_shared_aiohttp_session():
-    """Initialize shared aiohttp session for connection reuse."""
+    """Initialize shared aiohttp session for connection reuse with connection limits."""
     try:
         from aiohttp import ClientSession, TCPConnector
 
-        # Create connector with connection pooling settings optimized for long-lived connections
-        connector = TCPConnector(
-            limit=AIOHTTP_CONNECTOR_LIMIT,
-            keepalive_timeout=AIOHTTP_KEEPALIVE_TIMEOUT,
-            ttl_dns_cache=AIOHTTP_TTL_DNS_CACHE,
-            enable_cleanup_closed=AIOHTTP_NEEDS_CLEANUP_CLOSED,
-        )
-
+        connector_kwargs = {
+            "keepalive_timeout": AIOHTTP_KEEPALIVE_TIMEOUT,
+            "ttl_dns_cache": AIOHTTP_TTL_DNS_CACHE,
+            "enable_cleanup_closed": True,
+        }
+        if AIOHTTP_CONNECTOR_LIMIT > 0:
+            connector_kwargs["limit"] = AIOHTTP_CONNECTOR_LIMIT
+        if AIOHTTP_CONNECTOR_LIMIT_PER_HOST > 0:
+            connector_kwargs["limit_per_host"] = AIOHTTP_CONNECTOR_LIMIT_PER_HOST
+        
+        connector = TCPConnector(**connector_kwargs)
         session = ClientSession(connector=connector)
+        
         verbose_proxy_logger.info(
-            f"SESSION REUSE: Created shared aiohttp session for connection pooling (ID: {id(session)})"
+            f"SESSION REUSE: Created shared aiohttp session for connection pooling (ID: {id(session)}, "
+            f"limit={AIOHTTP_CONNECTOR_LIMIT}, limit_per_host={AIOHTTP_CONNECTOR_LIMIT_PER_HOST})"
         )
         return session
     except Exception as e: