diff --git a/litellm/caching/caching.py b/litellm/caching/caching.py
index 088922046e1..4a7ed6a76e9 100644
--- a/litellm/caching/caching.py
+++ b/litellm/caching/caching.py
@@ -393,13 +393,18 @@ class Cache:
                         continue  # ignore None params
                     param_value = kwargs[param]
                     cache_key += f"{param}: {param_value}"
-                elif kwargs[param] is not None and param not in _warned_dropped_cache_params:
+                elif (
+                    kwargs[param] is not None
+                    and param not in _warned_dropped_cache_params
+                    and len(_warned_dropped_cache_params) < 100
+                ):
                     _warned_dropped_cache_params.add(param)
                     verbose_logger.warning(
                         "litellm.cache: provider-specific param '%s' is excluded from the cache key by default, "
                         "so requests differing only in '%s' will return the same cached response. "
                         "Set litellm.enable_caching_on_provider_specific_optional_params=True to include it in the key.",
-                        param, param,
+                        param,
+                        param,
                     )
 
         if is_semantic_cache: