diff --git a/models/vllm/server.py b/models/vllm/server.py
index d7ae9b5..823b852 100644
--- a/models/vllm/server.py
+++ b/models/vllm/server.py
@@ -77,7 +77,8 @@ async def startup_event():
         dtype="auto",  # Auto-detect dtype
         download_dir="/workspace/huggingface_cache",  # Large disk
         trust_remote_code=True,  # Some models require this
-        enforce_eager=False,  # Use CUDA graphs for better performance
+        enforce_eager=True,  # Enable eager execution for proper streaming
+        disable_log_stats=True,  # Disable log stats for better streaming performance
     )
 
     # Create async engine