diff --git a/meshai/backends/base.py b/meshai/backends/base.py
index 6843e00..d50bb3f 100644
--- a/meshai/backends/base.py
+++ b/meshai/backends/base.py
@@ -12,7 +12,7 @@ class LLMBackend(ABC):
         self,
         messages: list[dict],
         system_prompt: str,
-        max_tokens: int = 300,
+        max_tokens: int = 8192,
         user_id: Optional[str] = None,
     ) -> str:
         """Generate a response from the LLM.
diff --git a/meshai/config.py b/meshai/config.py
index ce6c9af..d464386 100644
--- a/meshai/config.py
+++ b/meshai/config.py
@@ -94,6 +94,7 @@ class LLMConfig:
     base_url: str = "https://api.openai.com/v1"
     model: str = "gpt-4o-mini"
     timeout: int = 30
+    max_response_tokens: int = 8192  # Let LLM generate full responses; chunker handles size
 
     system_prompt: str = (
         "RESPONSE RULES:\n"
diff --git a/meshai/router.py b/meshai/router.py
index 4c19495..4cb0f9e 100644
--- a/meshai/router.py
+++ b/meshai/router.py
@@ -643,7 +643,7 @@ class MessageRouter:
             response = await self.llm.generate(
                 messages=history,
                 system_prompt=system_prompt,
-                max_tokens=500,
+                max_tokens=self.config.llm.max_response_tokens,
             )
         except asyncio.TimeoutError:
             logger.error("LLM request timed out")