diff --git a/nanobot/providers/litellm_provider.py b/nanobot/providers/litellm_provider.py
index edeb5c6..58c9ac2 100644
--- a/nanobot/providers/litellm_provider.py
+++ b/nanobot/providers/litellm_provider.py
@@ -111,7 +111,7 @@ class LiteLLMProvider(LLMProvider):
     def _supports_cache_control(self, model: str) -> bool:
         """Return True when the provider supports cache_control on content blocks."""
         if self._gateway is not None:
-            return False
+            return self._gateway.supports_prompt_caching
         spec = find_by_model(model)
         return spec is not None and spec.supports_prompt_caching
 
diff --git a/nanobot/providers/registry.py b/nanobot/providers/registry.py
index 445d977..ecf092f 100644
--- a/nanobot/providers/registry.py
+++ b/nanobot/providers/registry.py
@@ -100,6 +100,7 @@ PROVIDERS: tuple[ProviderSpec, ...] = (
         default_api_base="https://openrouter.ai/api/v1",
         strip_model_prefix=False,
         model_overrides=(),
+        supports_prompt_caching=True,
     ),
 
     # AiHubMix: global gateway, OpenAI-compatible interface.