diff --git a/litellm/__init__.py b/litellm/__init__.py
index c8df4394a06..9e83e7efd88 100644
--- a/litellm/__init__.py
+++ b/litellm/__init__.py
@@ -377,7 +377,9 @@ prompt_name_config_map: Dict[str, PromptSpec] = {}
 enable_preview_features: bool = False
 return_response_headers: bool = False  # get response headers from LLM Api providers - example x-remaining-requests,
 enable_json_schema_validation: bool = False
-enable_model_config_credential_overrides: bool = False
+enable_model_config_credential_overrides: bool = (
+    os.getenv("LITELLM_ENABLE_MODEL_CONFIG_CREDENTIAL_OVERRIDES", "false").lower() == "true"
+)
 enable_key_alias_format_validation: bool = (
     False  # opt-in validation of key_alias format on /key/generate and /key/update
 )
diff --git a/litellm/proxy/litellm_pre_call_utils.py b/litellm/proxy/litellm_pre_call_utils.py
index 90bba82aa84..234fe1b55dc 100644
--- a/litellm/proxy/litellm_pre_call_utils.py
+++ b/litellm/proxy/litellm_pre_call_utils.py
@@ -2667,7 +2667,6 @@ def _apply_credential_overrides_from_model_config(
     5. Team default override (defaultconfig)
     6. Deployment default (no action needed)
     """
-    # Feature flag gate — disabled by default, opt in with litellm.enable_model_config_credential_overrides = True
     if not litellm.enable_model_config_credential_overrides:
         return
 
diff --git a/tests/test_litellm/proxy/test_litellm_pre_call_utils.py b/tests/test_litellm/proxy/test_litellm_pre_call_utils.py
index 0d4b9e8d21f..b6454533972 100644
--- a/tests/test_litellm/proxy/test_litellm_pre_call_utils.py
+++ b/tests/test_litellm/proxy/test_litellm_pre_call_utils.py
@@ -2,9 +2,13 @@ import asyncio
 import copy
 import json
 import os
+import subprocess
+import sys
+import textwrap
 import time
 from datetime import datetime, timezone
 from types import SimpleNamespace
+from typing import Final
 from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
@@ -5152,6 +5156,36 @@ def test_apply_overrides_feature_flag_disabled_by_default():
     assert "api_key" not in data
 
 
+@pytest.mark.parametrize(
+    "env_value, expected",
+    [("true", True), ("True", True), ("false", False), (None, False)],
+)
+def test_credential_overrides_flag_reads_env_var_at_import(env_value: str | None, expected: bool):
+    """LITELLM_ENABLE_MODEL_CONFIG_CREDENTIAL_OVERRIDES is read at import, so check it in a fresh interpreter."""
+    base_env: Final = {k: v for k, v in os.environ.items() if k != "LITELLM_ENABLE_MODEL_CONFIG_CREDENTIAL_OVERRIDES"}
+    env: Final = (
+        base_env if env_value is None else {**base_env, "LITELLM_ENABLE_MODEL_CONFIG_CREDENTIAL_OVERRIDES": env_value}
+    )
+    result: Final = subprocess.run(
+        [
+            sys.executable,
+            "-c",
+            textwrap.dedent(
+                """
+                import litellm
+                print(litellm.enable_model_config_credential_overrides)
+                """
+            ),
+        ],
+        capture_output=True,
+        text=True,
+        env=env,
+        timeout=180,
+    )
+    assert result.returncode == 0, result.stderr
+    assert result.stdout.strip() == str(expected)
+
+
 def test_extract_credential_provider_hint_prefers_exact_match():
     """Provider hint selects the correct provider in a multi-provider entry."""
     entry = {