diff --git a/litellm/integrations/prometheus.py b/litellm/integrations/prometheus.py
index 1a899cf0e83..d9e57ee7cee 100644
--- a/litellm/integrations/prometheus.py
+++ b/litellm/integrations/prometheus.py
@@ -266,19 +266,6 @@ class PrometheusLogger(CustomLogger):
             # LiteLLM Virtual API KEY metrics
             ########################################
 
-            # Auth DB load diagnostic: count direct combined_view SQL queries.
-            # Each increment means a virtual-key cache miss that hit the DB.
-            # Useful for validating that enable_redis_auth_cache is working.
-            self.litellm_auth_combined_view_queries_total = self._counter_factory(
-                "litellm_auth_combined_view_queries_total",
-                "Number of times the combined_view SQL query was issued for virtual-key auth. "
-                "Each count is a cache miss that hit the database. Use to validate "
-                "enable_redis_auth_cache is reducing DB load.",
-                labelnames=self.get_labels_for_metric(
-                    "litellm_auth_combined_view_queries_total"
-                ),
-            )
-
             # Remaining MODEL RPM limit for API Key
             self.litellm_remaining_api_key_requests_for_model = self._gauge_factory(
                 "litellm_remaining_api_key_requests_for_model",
diff --git a/litellm/proxy/auth/auth_checks.py b/litellm/proxy/auth/auth_checks.py
index 9266a917c5f..113a8f538c0 100644
--- a/litellm/proxy/auth/auth_checks.py
+++ b/litellm/proxy/auth/auth_checks.py
@@ -59,7 +59,6 @@ from litellm.proxy._types import (
     SpecialModelNames,
     UserAPIKeyAuth,
 )
-from litellm.proxy.auth.auth_metrics import AuthMetrics
 from litellm.proxy.auth.route_checks import RouteChecks
 from litellm.proxy.db.exception_handler import PrismaDBExceptionHandler
 from litellm.proxy.guardrails.tool_name_extraction import (
@@ -2262,7 +2261,6 @@ async def _fetch_key_object_from_db_with_reconnect(
     Fetch key object from DB and retry once if a DB connection error can be healed.
     """
     try:
-        AuthMetrics.inc_combined_view_query(hashed_token)
         return await prisma_client.get_data(
             token=hashed_token,
             table_name="combined_view",
diff --git a/litellm/proxy/auth/auth_metrics.py b/litellm/proxy/auth/auth_metrics.py
deleted file mode 100644
index 7284ba42736..00000000000
--- a/litellm/proxy/auth/auth_metrics.py
+++ /dev/null
@@ -1,60 +0,0 @@
-"""
-Prometheus metric helpers for the auth layer.
-
-All metrics are thin wrappers around the shared ``PrometheusLogger`` instance so
-that every counter follows the same registration path (``_counter_factory``,
-label-filter config) as the rest of LiteLLM's metrics.
-
-Usage::
-
-    from litellm.proxy.auth.auth_metrics import AuthMetrics
-
-    AuthMetrics.inc_combined_view_query(hashed_token="sk-xxx")
-"""
-
-from litellm._logging import verbose_proxy_logger
-from litellm.types.integrations.prometheus import UserAPIKeyLabelValues
-
-
-class AuthMetrics:
-    """Static helpers for incrementing auth-layer Prometheus counters."""
-
-    @staticmethod
-    def _get_prom():
-        """Return the active PrometheusLogger, or None if Prometheus is not configured."""
-        try:
-            from litellm.router_utils.cooldown_callbacks import (
-                _get_prometheus_logger_from_callbacks,
-            )
-
-            return _get_prometheus_logger_from_callbacks()
-        except Exception:
-            return None
-
-    @staticmethod
-    def inc_combined_view_query(hashed_token: str) -> None:
-        """
-        Increment ``litellm_auth_combined_view_queries_total``.
-
-        Called once per virtual-key DB lookup (combined_view query). Each
-        increment represents a cache miss that hit the database — use this to
-        validate that ``enable_redis_auth_cache`` is reducing DB load.
-        """
-        try:
-            prom = AuthMetrics._get_prom()
-            if prom is not None:
-                # Counter labelnames include ``hashed_api_key`` plus any
-                # ``custom_prometheus_metadata_labels`` / ``custom_prometheus_tags``
-                # (see ``PrometheusMetricLabels.get_labels``). Use the same
-                # ``_inc_labeled_counter`` + ``prometheus_label_factory`` path as
-                # other metrics so label cardinality always matches registration.
-                prom._inc_labeled_counter(
-                    prom.litellm_auth_combined_view_queries_total,
-                    "litellm_auth_combined_view_queries_total",
-                    UserAPIKeyLabelValues(hashed_api_key=hashed_token),
-                )
-        except Exception as e:
-            verbose_proxy_logger.debug(
-                "AuthMetrics.inc_combined_view_query: failed to increment counter: %s",
-                e,
-            )
diff --git a/litellm/proxy/common_utils/cache_coordinator.py b/litellm/proxy/common_utils/cache_coordinator.py
index fc46ab6d78e..abb0402d3b9 100644
--- a/litellm/proxy/common_utils/cache_coordinator.py
+++ b/litellm/proxy/common_utils/cache_coordinator.py
@@ -153,7 +153,7 @@ class EventDrivenCacheCoordinator:
                 elapsed_ms,
                 value,
             )
-            
+
         await cache.async_set_cache(key=cache_key, value=value)
         if self._log_prefix:
             verbose_proxy_logger.debug("%s Result cached", self._log_prefix)
diff --git a/litellm/types/integrations/prometheus.py b/litellm/types/integrations/prometheus.py
index a4573535f72..43a287f29bc 100644
--- a/litellm/types/integrations/prometheus.py
+++ b/litellm/types/integrations/prometheus.py
@@ -228,8 +228,6 @@ DEFINED_PROMETHEUS_METRICS = Literal[
     "litellm_guardrail_latency_seconds",
     "litellm_guardrail_errors_total",
     "litellm_guardrail_requests_total",
-    # Auth DB diagnostic metrics
-    "litellm_auth_combined_view_queries_total",
     # Cache metrics
     "litellm_cache_hits_metric",
     "litellm_cache_misses_metric",
@@ -309,11 +307,6 @@ class PrometheusMetricLabels:
     litellm_guardrail_errors_total: List[str] = []
     litellm_guardrail_requests_total: List[str] = []
 
-    # Auth DB diagnostic - label by key so you can see which virtual key causes DB hits
-    litellm_auth_combined_view_queries_total = [
-        UserAPIKeyLabelNames.API_KEY_HASH.value,
-    ]
-
     litellm_proxy_total_requests_metric = [
         UserAPIKeyLabelNames.END_USER.value,
         UserAPIKeyLabelNames.API_KEY_HASH.value,