fix slow response

app.py

@@ -28,6 +28,7 @@ model.to(device)
 model.eval()
 
 # Print device and memory info for debugging
+print(f"---------- Info -----------")
 print(f"Device: {device}")
 print(f"Memory allocated: {torch.cuda.memory_allocated(device)/1e9:.2f} GB" if torch.cuda.is_available() else "CPU only")
 
@@ -39,6 +40,7 @@ def print_system_resources():
 
 # Call before generation
 print_system_resources()
+print(f"--------------------------")
 
 def generate_text(prompt, max_length=50, temperature=1.0):
     try: